preparation.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "confident-executive",
   "metadata": {},
   "source": [
    "# INF161: Data Science:  Course Project\n",
    "### *Predicting Soccer Scores for the Norwegian Women's Soccer Championships (Toppserien)*\n",
    "\n",
    "> #### by [Sebastian Einar Røkholt](https://www.linkedin.com/in/sebastianrokholt/)\n",
    ">-------------------------------------------\n",
    "> *Autumn 2021* \n",
    "----------------------------------------------------------------------------\n",
    "\n",
    "<a id=\"top\"></a> \n",
    "\n",
    "<h2> Part 1: Preparations</h2>\n",
    "    \n",
    "#### *Notebook Index:*\n",
    "1. [**Initial Inspection and Handling Missing Values**](#inspection)<br>\n",
    "2. [**Feature Selection**](#feature-selection) <br>\n",
    "3. [**Feature Engineering**](#feature-engineering) <br>\n",
    "    3.1 [*Adding the opponents' matches to the dataset*](#opponent-matches) <br>\n",
    "    3.2 [*Converting the target feature*](#conversion) <br>\n",
    "    3.3 [*Preparing the data from the 2020 season*](#preparing-2020-data) <br>\n",
    "    3.4 [*Visualising the distribution of scores*](#visualising-scores) <br>\n",
    "    3.5 [*Adding game season and game number in season*](#game-season) <br>\n",
    "    3.6 [*Adding the average historical score difference*](#ahsd) <br>\n",
    "    3.7 [*Adding the average number of goals per 90 min last season*](#avg-nr-goals) <br>\n",
    "    3.8 [*Adding the teams' average age*](#team-avg-age) <br>\n",
    "    3.9 [*Adding a feature to distinguish top-tier teams*](#top-tier-team) <br>\n",
    "    3.10 [*One Hot Encoding categorical features*](#ohe) <br>\n",
    "4. [**Saving the Cleaned Data**](#saving) <br>\n",
    "<img src=\"broom.png\" alt=\"A broom\" align=\"left\" width=\"150\" height=\"150\"/>"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b33c6b70",
   "metadata": {},
   "source": [
    "The first part of this machine learning project will consist of data inspection, handling missing values, feature selection and feature engineering. The primary aim of Part 1 is to produce a cleaned dataset to be used for training and evaluating machine learning models. The secondary aim is to engineer features that are useful for predicting the outcome of a soccer match. In order to achieve this, it is important that I visualise the data in a way that allows me to investigate interesting relationships and patterns. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "jewish-massachusetts",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Importing notebook dependencies\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "from collections import defaultdict\n",
    "\n",
    "\n",
    "# Setting options\n",
    "pd.set_option(\"display.max_columns\", 50)\n",
    "pd.set_option(\"display.max_rows\", 500)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "described-balloon",
   "metadata": {},
   "source": [
    "## <a id=\"inspection\"></a>1) Initial Inspection and Handling Missing Values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "fourth-jacket",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Wk</th>\n",
       "      <th>Day</th>\n",
       "      <th>Date</th>\n",
       "      <th>Time</th>\n",
       "      <th>Home</th>\n",
       "      <th>Score</th>\n",
       "      <th>Away</th>\n",
       "      <th>Attendance</th>\n",
       "      <th>Venue</th>\n",
       "      <th>Referee</th>\n",
       "      <th>Match Report</th>\n",
       "      <th>Notes</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.0</td>\n",
       "      <td>Mon</td>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>13:00</td>\n",
       "      <td>Medkila</td>\n",
       "      <td>1–1</td>\n",
       "      <td>Sandviken</td>\n",
       "      <td>307.0</td>\n",
       "      <td>Harstad Stadion</td>\n",
       "      <td>Sarah Fatemeh Zangeneh</td>\n",
       "      <td>Match Report</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1.0</td>\n",
       "      <td>Mon</td>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>14:00</td>\n",
       "      <td>Avaldsnes</td>\n",
       "      <td>2–1</td>\n",
       "      <td>Vålerenga</td>\n",
       "      <td>346.0</td>\n",
       "      <td>Avaldsnes Idrettssenter</td>\n",
       "      <td>Marit Folstad</td>\n",
       "      <td>Match Report</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1.0</td>\n",
       "      <td>Mon</td>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>15:00</td>\n",
       "      <td>Grand Bodø</td>\n",
       "      <td>2–2</td>\n",
       "      <td>Arna-Bjørnar</td>\n",
       "      <td>290.0</td>\n",
       "      <td>Nordlandshallen</td>\n",
       "      <td>Henrikke Holm Nervik</td>\n",
       "      <td>Match Report</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1.0</td>\n",
       "      <td>Mon</td>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>15:00</td>\n",
       "      <td>Røa</td>\n",
       "      <td>1–2</td>\n",
       "      <td>Stabæk</td>\n",
       "      <td>348.0</td>\n",
       "      <td>Røabanen</td>\n",
       "      <td>Emilie Dokset</td>\n",
       "      <td>Match Report</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1.0</td>\n",
       "      <td>Mon</td>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>15:00</td>\n",
       "      <td>LSK Kvinner</td>\n",
       "      <td>6–1</td>\n",
       "      <td>Klepp</td>\n",
       "      <td>280.0</td>\n",
       "      <td>LSK-hallen</td>\n",
       "      <td>Ingvild Aarland</td>\n",
       "      <td>Match Report</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    Wk  Day        Date   Time         Home Score          Away  Attendance  \\\n",
       "0  1.0  Mon  2017-04-17  13:00      Medkila   1–1     Sandviken       307.0   \n",
       "1  1.0  Mon  2017-04-17  14:00    Avaldsnes   2–1     Vålerenga       346.0   \n",
       "2  1.0  Mon  2017-04-17  15:00   Grand Bodø   2–2  Arna-Bjørnar       290.0   \n",
       "3  1.0  Mon  2017-04-17  15:00          Røa   1–2        Stabæk       348.0   \n",
       "4  1.0  Mon  2017-04-17  15:00  LSK Kvinner   6–1         Klepp       280.0   \n",
       "\n",
       "                     Venue                 Referee  Match Report  Notes  \n",
       "0          Harstad Stadion  Sarah Fatemeh Zangeneh  Match Report    NaN  \n",
       "1  Avaldsnes Idrettssenter           Marit Folstad  Match Report    NaN  \n",
       "2          Nordlandshallen    Henrikke Holm Nervik  Match Report    NaN  \n",
       "3                 Røabanen           Emilie Dokset  Match Report    NaN  \n",
       "4               LSK-hallen         Ingvild Aarland  Match Report    NaN  "
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Loading the games dataset\n",
    "games_2017_df = pd.read_html(\"Raw Data/2017/games.xls\", encoding=\"UTF-8\")[0]\n",
    "games_2018_df = pd.read_html(\"Raw Data/2018/games.xls\", encoding=\"UTF-8\")[0]\n",
    "games_2019_df = pd.read_html(\"Raw Data/2019/games.xls\", encoding=\"UTF-8\")[0]\n",
    "\n",
    "games_2017_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "induced-victor",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Wk</th>\n",
       "      <th>Day</th>\n",
       "      <th>Date</th>\n",
       "      <th>Time</th>\n",
       "      <th>Home</th>\n",
       "      <th>Score</th>\n",
       "      <th>Away</th>\n",
       "      <th>Attendance</th>\n",
       "      <th>Venue</th>\n",
       "      <th>Referee</th>\n",
       "      <th>Match Report</th>\n",
       "      <th>Notes</th>\n",
       "      <th>Round</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.0</td>\n",
       "      <td>Mon</td>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>13:00</td>\n",
       "      <td>Medkila</td>\n",
       "      <td>1–1</td>\n",
       "      <td>Sandviken</td>\n",
       "      <td>307.0</td>\n",
       "      <td>Harstad Stadion</td>\n",
       "      <td>Sarah Fatemeh Zangeneh</td>\n",
       "      <td>Match Report</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1.0</td>\n",
       "      <td>Mon</td>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>14:00</td>\n",
       "      <td>Avaldsnes</td>\n",
       "      <td>2–1</td>\n",
       "      <td>Vålerenga</td>\n",
       "      <td>346.0</td>\n",
       "      <td>Avaldsnes Idrettssenter</td>\n",
       "      <td>Marit Folstad</td>\n",
       "      <td>Match Report</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1.0</td>\n",
       "      <td>Mon</td>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>15:00</td>\n",
       "      <td>Grand Bodø</td>\n",
       "      <td>2–2</td>\n",
       "      <td>Arna-Bjørnar</td>\n",
       "      <td>290.0</td>\n",
       "      <td>Nordlandshallen</td>\n",
       "      <td>Henrikke Holm Nervik</td>\n",
       "      <td>Match Report</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1.0</td>\n",
       "      <td>Mon</td>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>15:00</td>\n",
       "      <td>Røa</td>\n",
       "      <td>1–2</td>\n",
       "      <td>Stabæk</td>\n",
       "      <td>348.0</td>\n",
       "      <td>Røabanen</td>\n",
       "      <td>Emilie Dokset</td>\n",
       "      <td>Match Report</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1.0</td>\n",
       "      <td>Mon</td>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>15:00</td>\n",
       "      <td>LSK Kvinner</td>\n",
       "      <td>6–1</td>\n",
       "      <td>Klepp</td>\n",
       "      <td>280.0</td>\n",
       "      <td>LSK-hallen</td>\n",
       "      <td>Ingvild Aarland</td>\n",
       "      <td>Match Report</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    Wk  Day        Date   Time         Home Score          Away  Attendance  \\\n",
       "0  1.0  Mon  2017-04-17  13:00      Medkila   1–1     Sandviken       307.0   \n",
       "1  1.0  Mon  2017-04-17  14:00    Avaldsnes   2–1     Vålerenga       346.0   \n",
       "2  1.0  Mon  2017-04-17  15:00   Grand Bodø   2–2  Arna-Bjørnar       290.0   \n",
       "3  1.0  Mon  2017-04-17  15:00          Røa   1–2        Stabæk       348.0   \n",
       "4  1.0  Mon  2017-04-17  15:00  LSK Kvinner   6–1         Klepp       280.0   \n",
       "\n",
       "                     Venue                 Referee  Match Report Notes Round  \n",
       "0          Harstad Stadion  Sarah Fatemeh Zangeneh  Match Report   NaN   NaN  \n",
       "1  Avaldsnes Idrettssenter           Marit Folstad  Match Report   NaN   NaN  \n",
       "2          Nordlandshallen    Henrikke Holm Nervik  Match Report   NaN   NaN  \n",
       "3                 Røabanen           Emilie Dokset  Match Report   NaN   NaN  \n",
       "4               LSK-hallen         Ingvild Aarland  Match Report   NaN   NaN  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Concatenating the games data for 2017, 2018 and 2019 to one dataframe\n",
    "games_df = pd.concat([games_2017_df, games_2018_df, games_2019_df])\n",
    "games_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "eccea64d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Shape of games_df: (479, 13)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "Wk              float64\n",
       "Day              object\n",
       "Date             object\n",
       "Time             object\n",
       "Home             object\n",
       "Score            object\n",
       "Away             object\n",
       "Attendance      float64\n",
       "Venue            object\n",
       "Referee          object\n",
       "Match Report     object\n",
       "Notes            object\n",
       "Round            object\n",
       "dtype: object"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Initial exploration of the dataset\n",
    "# Printing the shape\n",
    "print(f\"Shape of games_df: {games_df.shape}\")\n",
    "\n",
    "# Printing the data types for all columns\n",
    "games_df.dtypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "da2183de",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Wk               83\n",
       "Day              81\n",
       "Date             81\n",
       "Time             81\n",
       "Home             81\n",
       "Score            81\n",
       "Away             81\n",
       "Attendance       86\n",
       "Venue            81\n",
       "Referee          85\n",
       "Match Report     81\n",
       "Notes           477\n",
       "Round           345\n",
       "dtype: int64"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Printing the number of missing values for each column\n",
    "games_df.isna().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "53ff0506",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "There are 81 rows where all values are missing.\n"
     ]
    }
   ],
   "source": [
    "# Counting rows where all values are missing\n",
    "games_df_no_nanrows = games_df.dropna(how=\"all\")\n",
    "print(f\"There are {games_df.shape[0]- games_df_no_nanrows.shape[0]} rows where all values are missing.\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "ce601e1f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "There are 81 duplicated rows in the dataset\n"
     ]
    }
   ],
   "source": [
    "# Checking for duplicate rows\n",
    "print(f\"There are {len(games_df[games_df.duplicated(keep=False)])} duplicated rows in the dataset\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "12c27e6c",
   "metadata": {},
   "source": [
    "There are a lot of missing values in the dataset, and some features consist almost entirely of missing values. Luckily there are only a few features which are suitable for prediction. There are also 81 rows that only contain missing values, which make up the bulk of the missing values in the dataset. There are no duplicate rows except for the ones where all values are missing. Removing unwanted features and empty rows, we are left with no missing values in the remaining features."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "optical-attention",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Remaining missing values:\n",
      " Wk              2\n",
      "Day             0\n",
      "Date            0\n",
      "Time            0\n",
      "Home            0\n",
      "Score           0\n",
      "Away            0\n",
      "Attendance      5\n",
      "Venue           0\n",
      "Referee         4\n",
      "Match Report    0\n",
      "dtype: int64\n"
     ]
    }
   ],
   "source": [
    "# Removing columns with >300 NaN values\n",
    "games_df.drop(columns=[\"Notes\", \"Round\"], inplace=True)\n",
    "games_df.head()\n",
    "\n",
    "# Removing rows with only NaN values\n",
    "games_df.dropna(axis=0, how=\"all\", inplace=True)\n",
    "\n",
    "print(\"Remaining missing values:\\n\", games_df.isna().sum())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fe1d6089",
   "metadata": {},
   "source": [
    "[*back to top*](#top)\n",
    "## <a id=\"feature-selection\"></a>2) Feature Selection"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d5121c47",
   "metadata": {},
   "source": [
    "I will now look at how the remaining data is represented and which features are suitable for predicting match outcomes. <br>\n",
    "Since we the \"Match Report\" feature only contains the string \"match report\" repeated for all rows, I can safely remove it from the data. Other features are simply unsuitable for predicting soccer scores, at least without supplementing the dataset with more data from external resources. For example, the \"Wk\", \"Day\" and \"Time\" features could help improve predictions if I added weather data for each game. However, I will be sticking to the original dataset, so I will remove these as I deem it unlikely that the day of the week or time of day will consistently impact the performance of some teams more than others. If there is an effect, say that teams play better earlier in the day because they are well rested, it is likely that the effect will be equal for all players on both teams. \n",
    "\n",
    "Since a refereees job is to be as objective as possible - though this is sometimes disputed by fans - I will remove the feature containing the name of the referee for each match. Furthermore, though the attendance for each match might slightly affect the performance of the teams playing, I will remove the \"Attendance\" feature simply because I don't have any data for the expected attendance (tickets sold) for the games I will try to predict for. \n",
    "\n",
    "The resulting dataframe should have no missing values, as I am removing the remainding features with NaN values. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "e4fe7474",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Removing column 'Match Report'\n",
    "games_df.drop(columns=\"Match Report\", inplace=True)\n",
    "\n",
    "# Removing other columns that are unsuitable for predicting scores \n",
    "games_df.drop(columns=[\"Wk\", \"Day\", \"Time\", \"Attendance\", \"Referee\"], inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "50318102",
   "metadata": {},
   "source": [
    "I am left with the 5 features from the original dataset:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "7710d346",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Date</th>\n",
       "      <th>Home</th>\n",
       "      <th>Score</th>\n",
       "      <th>Away</th>\n",
       "      <th>Venue</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>Medkila</td>\n",
       "      <td>1–1</td>\n",
       "      <td>Sandviken</td>\n",
       "      <td>Harstad Stadion</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>Avaldsnes</td>\n",
       "      <td>2–1</td>\n",
       "      <td>Vålerenga</td>\n",
       "      <td>Avaldsnes Idrettssenter</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>Grand Bodø</td>\n",
       "      <td>2–2</td>\n",
       "      <td>Arna-Bjørnar</td>\n",
       "      <td>Nordlandshallen</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>Røa</td>\n",
       "      <td>1–2</td>\n",
       "      <td>Stabæk</td>\n",
       "      <td>Røabanen</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>LSK Kvinner</td>\n",
       "      <td>6–1</td>\n",
       "      <td>Klepp</td>\n",
       "      <td>LSK-hallen</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         Date         Home Score          Away                    Venue\n",
       "0  2017-04-17      Medkila   1–1     Sandviken          Harstad Stadion\n",
       "1  2017-04-17    Avaldsnes   2–1     Vålerenga  Avaldsnes Idrettssenter\n",
       "2  2017-04-17   Grand Bodø   2–2  Arna-Bjørnar          Nordlandshallen\n",
       "3  2017-04-17          Røa   1–2        Stabæk                 Røabanen\n",
       "4  2017-04-17  LSK Kvinner   6–1         Klepp               LSK-hallen"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Displaying the first 5 rows\n",
    "games_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ef586289",
   "metadata": {},
   "source": [
    "[*back to top*](#top)\n",
    "## <a id=\"feature-engineering\"></a>3) Feature Engineering"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "694030cb",
   "metadata": {},
   "source": [
    "### <a id=\"opponent-matches\"></a>3.1) Adding the opponents' matches to the dataset"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2dbeb6fa",
   "metadata": {},
   "source": [
    "There are still a few things that need to be fixed in terms of how the data is represented. \n",
    "\n",
    "A larger issue is that I ultimately want to create a machine learning model that takes as its input the name of a team, the name of it's opponent, and whether the team i playing at home or away. It should then predict the probability of a win, tie or loss. The problem here is that each match is only present in the dataset once, even though there are always two teams playing and therefore two different situations that can be predicted for, i.e: Team A is playing a home game against opponent B at, while the other situation is that Team B is playing an away game against opponent A. \n",
    "\n",
    "The outcome of these situations are different, even though they occur simultaneously (and the results are the inverse of each other). To account for this, I am going to rename the features into something more adequate, duplicate the dataframe, switch the teams and the scores so that each match is represented twice and then append the second dataframe to the end of the first. I will then resort the new dataframe ascending by date. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "7f8eab4b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Date</th>\n",
       "      <th>Team</th>\n",
       "      <th>Score</th>\n",
       "      <th>Opponent</th>\n",
       "      <th>Venue</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>Medkila</td>\n",
       "      <td>1–1</td>\n",
       "      <td>Sandviken</td>\n",
       "      <td>Home</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>Avaldsnes</td>\n",
       "      <td>2–1</td>\n",
       "      <td>Vålerenga</td>\n",
       "      <td>Home</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>Grand Bodø</td>\n",
       "      <td>2–2</td>\n",
       "      <td>Arna-Bjørnar</td>\n",
       "      <td>Home</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>Røa</td>\n",
       "      <td>1–2</td>\n",
       "      <td>Stabæk</td>\n",
       "      <td>Home</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>LSK Kvinner</td>\n",
       "      <td>6–1</td>\n",
       "      <td>Klepp</td>\n",
       "      <td>Home</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         Date         Team Score      Opponent Venue\n",
       "0  2017-04-17      Medkila   1–1     Sandviken  Home\n",
       "1  2017-04-17    Avaldsnes   2–1     Vålerenga  Home\n",
       "2  2017-04-17   Grand Bodø   2–2  Arna-Bjørnar  Home\n",
       "3  2017-04-17          Røa   1–2        Stabæk  Home\n",
       "4  2017-04-17  LSK Kvinner   6–1         Klepp  Home"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Converting all venue locations to \"Home\"\n",
    "games_df[\"Venue\"] = \"Home\"\n",
    "\n",
    "# Renaming dataframe to reflect that it only contains data about the home team\n",
    "home_games_df = games_df\n",
    "\n",
    "# Renaming feature \"Home\" to \"Team\"\n",
    "home_games_df.rename(columns={\"Home\": \"Team\", \"Away\":\"Opponent\"}, inplace=True)\n",
    "\n",
    "home_games_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b834cab7",
   "metadata": {},
   "source": [
    "I will now continue by extending the games dataframe with the opposite configuration of teams and opponents so that every match is represented twice."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "boring-server",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Date</th>\n",
       "      <th>Team</th>\n",
       "      <th>Opponent</th>\n",
       "      <th>Venue</th>\n",
       "      <th>Score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>Sandviken</td>\n",
       "      <td>Medkila</td>\n",
       "      <td>Away</td>\n",
       "      <td>1–1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>Vålerenga</td>\n",
       "      <td>Avaldsnes</td>\n",
       "      <td>Away</td>\n",
       "      <td>1–2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>Arna-Bjørnar</td>\n",
       "      <td>Grand Bodø</td>\n",
       "      <td>Away</td>\n",
       "      <td>2–2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>Stabæk</td>\n",
       "      <td>Røa</td>\n",
       "      <td>Away</td>\n",
       "      <td>2–1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>Klepp</td>\n",
       "      <td>LSK Kvinner</td>\n",
       "      <td>Away</td>\n",
       "      <td>1–6</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         Date          Team     Opponent Venue Score\n",
       "0  2017-04-17     Sandviken      Medkila  Away   1–1\n",
       "1  2017-04-17     Vålerenga    Avaldsnes  Away   1–2\n",
       "2  2017-04-17  Arna-Bjørnar   Grand Bodø  Away   2–2\n",
       "3  2017-04-17        Stabæk          Røa  Away   2–1\n",
       "4  2017-04-17         Klepp  LSK Kvinner  Away   1–6"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Creating a new df with all of the away games\n",
    "# Copying the df with the home games\n",
    "away_games_df = home_games_df.copy(deep=True)\n",
    "\n",
    "# All of these matches are away games\n",
    "away_games_df[\"Venue\"] = \"Away\"\n",
    "\n",
    "# Renaming so that the home team becomes the opposing team and vice versa\n",
    "away_games_df.rename(columns={\"Team\": \"Opponent\", \"Opponent\":\"Team\"}, inplace=True)\n",
    "\n",
    "# Reversing the score\n",
    "away_games_df[\"Score\"] = away_games_df[\"Score\"].apply(lambda x: str(x)[::-1])\n",
    "\n",
    "# Reordering columns\n",
    "away_games_df = away_games_df[[\"Date\", \"Team\", \"Opponent\", \"Venue\", \"Score\"]]\n",
    "\n",
    "# Displaying the new dataframe with only the away games\n",
    "display(away_games_df.head())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "36ecdce5",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Date</th>\n",
       "      <th>Team</th>\n",
       "      <th>Score</th>\n",
       "      <th>Opponent</th>\n",
       "      <th>Venue</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>Medkila</td>\n",
       "      <td>1–1</td>\n",
       "      <td>Sandviken</td>\n",
       "      <td>Home</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>Arna-Bjørnar</td>\n",
       "      <td>2–2</td>\n",
       "      <td>Grand Bodø</td>\n",
       "      <td>Away</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>Stabæk</td>\n",
       "      <td>2–1</td>\n",
       "      <td>Røa</td>\n",
       "      <td>Away</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>Klepp</td>\n",
       "      <td>1–6</td>\n",
       "      <td>LSK Kvinner</td>\n",
       "      <td>Away</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>Kolbotn</td>\n",
       "      <td>2–3</td>\n",
       "      <td>Trondheims-Ørn</td>\n",
       "      <td>Away</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>Vålerenga</td>\n",
       "      <td>1–2</td>\n",
       "      <td>Avaldsnes</td>\n",
       "      <td>Away</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>Sandviken</td>\n",
       "      <td>1–1</td>\n",
       "      <td>Medkila</td>\n",
       "      <td>Away</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>Trondheims-Ørn</td>\n",
       "      <td>3–2</td>\n",
       "      <td>Kolbotn</td>\n",
       "      <td>Home</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>LSK Kvinner</td>\n",
       "      <td>6–1</td>\n",
       "      <td>Klepp</td>\n",
       "      <td>Home</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>Røa</td>\n",
       "      <td>1–2</td>\n",
       "      <td>Stabæk</td>\n",
       "      <td>Home</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         Date            Team Score        Opponent Venue\n",
       "0  2017-04-17         Medkila   1–1       Sandviken  Home\n",
       "1  2017-04-17    Arna-Bjørnar   2–2      Grand Bodø  Away\n",
       "2  2017-04-17          Stabæk   2–1             Røa  Away\n",
       "3  2017-04-17           Klepp   1–6     LSK Kvinner  Away\n",
       "4  2017-04-17         Kolbotn   2–3  Trondheims-Ørn  Away\n",
       "5  2017-04-17       Vålerenga   1–2       Avaldsnes  Away\n",
       "6  2017-04-17       Sandviken   1–1         Medkila  Away\n",
       "7  2017-04-17  Trondheims-Ørn   3–2         Kolbotn  Home\n",
       "8  2017-04-17     LSK Kvinner   6–1           Klepp  Home\n",
       "9  2017-04-17             Røa   1–2          Stabæk  Home"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Appending the away games dataframe to the home games dataframe\n",
    "games_df = home_games_df.append(away_games_df, ignore_index=True)\n",
    "\n",
    "# Sorting games by dates, with most recent games at the top\n",
    "games_df.sort_values(by=\"Date\", axis=0, inplace=True)\n",
    "\n",
    "# Resetting the index\n",
    "games_df.reset_index(drop=True, inplace=True)\n",
    "\n",
    "games_df.head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "74d201e7",
   "metadata": {},
   "source": [
    "Finally, the target variable \"Score\" is represented as a string object. It needs to be split into two different dependent variables and converted to numerical values. "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b5ef12a5",
   "metadata": {},
   "source": [
    "### <a id=\"conversion\"></a>3.2) Converting the target feature"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "e9de9ab7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Date</th>\n",
       "      <th>Team</th>\n",
       "      <th>Opponent</th>\n",
       "      <th>Venue</th>\n",
       "      <th>y1: Team Score</th>\n",
       "      <th>y2: Opponent Score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>Medkila</td>\n",
       "      <td>Sandviken</td>\n",
       "      <td>Home</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>Arna-Bjørnar</td>\n",
       "      <td>Grand Bodø</td>\n",
       "      <td>Away</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>Stabæk</td>\n",
       "      <td>Røa</td>\n",
       "      <td>Away</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>Klepp</td>\n",
       "      <td>LSK Kvinner</td>\n",
       "      <td>Away</td>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>Kolbotn</td>\n",
       "      <td>Trondheims-Ørn</td>\n",
       "      <td>Away</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         Date          Team        Opponent Venue  y1: Team Score  \\\n",
       "0  2017-04-17       Medkila       Sandviken  Home               1   \n",
       "1  2017-04-17  Arna-Bjørnar      Grand Bodø  Away               2   \n",
       "2  2017-04-17        Stabæk             Røa  Away               2   \n",
       "3  2017-04-17         Klepp     LSK Kvinner  Away               1   \n",
       "4  2017-04-17       Kolbotn  Trondheims-Ørn  Away               2   \n",
       "\n",
       "   y2: Opponent Score  \n",
       "0                   1  \n",
       "1                   2  \n",
       "2                   1  \n",
       "3                   6  \n",
       "4                   3  "
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Representing target feature (\"Score\") as two dependent features suitable for multioutput regression\n",
    "games_df[\"y1: Team Score\"] = games_df[\"Score\"].apply(lambda x: int(x[0]))  # should be a regex, in case there are 10 goals or more! fix later\n",
    "games_df[\"y2: Opponent Score\"] = games_df[\"Score\"].apply(lambda x: int(x[-1]))\n",
    "games_df.drop([\"Score\"], axis=1, inplace=True)\n",
    "games_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4a8e8608",
   "metadata": {},
   "source": [
    "### <a id=\"preparing-2020-data\"></a>3.3) Preparing the 2020 games data"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e0a2be49",
   "metadata": {},
   "source": [
    "The test data needs to be on the same format as above."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "84a99587",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>ID</th>\n",
       "      <th>Date</th>\n",
       "      <th>Season</th>\n",
       "      <th>Team</th>\n",
       "      <th>Opponent</th>\n",
       "      <th>Venue</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>2020-08-02</td>\n",
       "      <td>2020</td>\n",
       "      <td>Trondheims-Ørn</td>\n",
       "      <td>Klepp</td>\n",
       "      <td>Home</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>2020-07-10</td>\n",
       "      <td>2020</td>\n",
       "      <td>Røa</td>\n",
       "      <td>Vålerenga</td>\n",
       "      <td>Away</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>2020-08-30</td>\n",
       "      <td>2020</td>\n",
       "      <td>Trondheims-Ørn</td>\n",
       "      <td>Arna-Bjørnar</td>\n",
       "      <td>Home</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>2020-08-26</td>\n",
       "      <td>2020</td>\n",
       "      <td>Vålerenga</td>\n",
       "      <td>Lyn</td>\n",
       "      <td>Home</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>2020-09-12</td>\n",
       "      <td>2020</td>\n",
       "      <td>Arna-Bjørnar</td>\n",
       "      <td>Kolbotn</td>\n",
       "      <td>Home</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   ID        Date  Season            Team      Opponent Venue\n",
       "0   0  2020-08-02    2020  Trondheims-Ørn         Klepp  Home\n",
       "1   1  2020-07-10    2020             Røa     Vålerenga  Away\n",
       "2   2  2020-08-30    2020  Trondheims-Ørn  Arna-Bjørnar  Home\n",
       "3   3  2020-08-26    2020       Vålerenga           Lyn  Home\n",
       "4   4  2020-09-12    2020    Arna-Bjørnar       Kolbotn  Home"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ID           int64\n",
      "Date        object\n",
      "Season       int64\n",
      "Team        object\n",
      "Opponent    object\n",
      "Venue       object\n",
      "dtype: object\n"
     ]
    }
   ],
   "source": [
    "games_2020_df = pd.read_csv(\"Raw Data/2020/test.csv\", encoding=\"UTF-8\")\n",
    "display(games_2020_df.head())\n",
    "\n",
    "print(games_2020_df.dtypes)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "c46d14a7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Date</th>\n",
       "      <th>Season</th>\n",
       "      <th>Team</th>\n",
       "      <th>Opponent</th>\n",
       "      <th>Venue</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>153</th>\n",
       "      <td>2020-07-03</td>\n",
       "      <td>2020</td>\n",
       "      <td>LSK Kvinner</td>\n",
       "      <td>Trondheims-Ørn</td>\n",
       "      <td>Away</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>172</th>\n",
       "      <td>2020-07-03</td>\n",
       "      <td>2020</td>\n",
       "      <td>Trondheims-Ørn</td>\n",
       "      <td>LSK Kvinner</td>\n",
       "      <td>Home</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>164</th>\n",
       "      <td>2020-07-04</td>\n",
       "      <td>2020</td>\n",
       "      <td>Sandviken</td>\n",
       "      <td>Vålerenga</td>\n",
       "      <td>Home</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>136</th>\n",
       "      <td>2020-07-04</td>\n",
       "      <td>2020</td>\n",
       "      <td>Røa</td>\n",
       "      <td>Arna-Bjørnar</td>\n",
       "      <td>Home</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>177</th>\n",
       "      <td>2020-07-04</td>\n",
       "      <td>2020</td>\n",
       "      <td>Avaldsnes</td>\n",
       "      <td>Lyn</td>\n",
       "      <td>Home</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>130</th>\n",
       "      <td>2020-07-04</td>\n",
       "      <td>2020</td>\n",
       "      <td>Lyn</td>\n",
       "      <td>Avaldsnes</td>\n",
       "      <td>Away</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>119</th>\n",
       "      <td>2020-07-04</td>\n",
       "      <td>2020</td>\n",
       "      <td>Vålerenga</td>\n",
       "      <td>Sandviken</td>\n",
       "      <td>Away</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>182</th>\n",
       "      <td>2020-07-04</td>\n",
       "      <td>2020</td>\n",
       "      <td>Arna-Bjørnar</td>\n",
       "      <td>Røa</td>\n",
       "      <td>Away</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>132</th>\n",
       "      <td>2020-07-05</td>\n",
       "      <td>2020</td>\n",
       "      <td>Klepp</td>\n",
       "      <td>Kolbotn</td>\n",
       "      <td>Home</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>161</th>\n",
       "      <td>2020-07-05</td>\n",
       "      <td>2020</td>\n",
       "      <td>Kolbotn</td>\n",
       "      <td>Klepp</td>\n",
       "      <td>Away</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          Date  Season            Team        Opponent Venue\n",
       "153 2020-07-03    2020     LSK Kvinner  Trondheims-Ørn  Away\n",
       "172 2020-07-03    2020  Trondheims-Ørn     LSK Kvinner  Home\n",
       "164 2020-07-04    2020       Sandviken       Vålerenga  Home\n",
       "136 2020-07-04    2020             Røa    Arna-Bjørnar  Home\n",
       "177 2020-07-04    2020       Avaldsnes             Lyn  Home\n",
       "130 2020-07-04    2020             Lyn       Avaldsnes  Away\n",
       "119 2020-07-04    2020       Vålerenga       Sandviken  Away\n",
       "182 2020-07-04    2020    Arna-Bjørnar             Røa  Away\n",
       "132 2020-07-05    2020           Klepp         Kolbotn  Home\n",
       "161 2020-07-05    2020         Kolbotn           Klepp  Away"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Remove feature \"ID\" as it is equal to the index, and can be reattached later\n",
    "game_ids = games_2020_df.ID\n",
    "games_2020_df = games_2020_df.drop(labels=\"ID\", axis=1)\n",
    "\n",
    "# Converting Date column from string to datetime\n",
    "games_2020_df[\"Date\"] = pd.to_datetime(games_2020_df[\"Date\"], yearfirst=True)\n",
    "\n",
    "# Ensure that each game occurs twice in the dataset, \n",
    "# once from the perspective of the home team and once for away\n",
    "# NB! 2020 data must still be unsorted so that we can re-attach ID's later\n",
    "sorted_games_2020_df = games_2020_df.sort_values(by=\"Date\", axis=0, inplace=False)\n",
    "sorted_games_2020_df.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "8a503edf",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Date</th>\n",
       "      <th>Team</th>\n",
       "      <th>Opponent</th>\n",
       "      <th>Venue</th>\n",
       "      <th>y1: Team Score</th>\n",
       "      <th>y2: Opponent Score</th>\n",
       "      <th>Season</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>174</th>\n",
       "      <td>2020-08-16 00:00:00</td>\n",
       "      <td>Kolbotn</td>\n",
       "      <td>Sandviken</td>\n",
       "      <td>Home</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2020.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>175</th>\n",
       "      <td>2020-12-06 00:00:00</td>\n",
       "      <td>Avaldsnes</td>\n",
       "      <td>Sandviken</td>\n",
       "      <td>Away</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2020.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>176</th>\n",
       "      <td>2020-10-03 00:00:00</td>\n",
       "      <td>Arna-Bjørnar</td>\n",
       "      <td>LSK Kvinner</td>\n",
       "      <td>Home</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2020.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>177</th>\n",
       "      <td>2020-07-04 00:00:00</td>\n",
       "      <td>Avaldsnes</td>\n",
       "      <td>Lyn</td>\n",
       "      <td>Home</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2020.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>178</th>\n",
       "      <td>2020-08-26 00:00:00</td>\n",
       "      <td>Arna-Bjørnar</td>\n",
       "      <td>Sandviken</td>\n",
       "      <td>Home</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2020.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>179</th>\n",
       "      <td>2020-09-27 00:00:00</td>\n",
       "      <td>Arna-Bjørnar</td>\n",
       "      <td>Sandviken</td>\n",
       "      <td>Away</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2020.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>180</th>\n",
       "      <td>2020-07-17 00:00:00</td>\n",
       "      <td>Røa</td>\n",
       "      <td>LSK Kvinner</td>\n",
       "      <td>Home</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2020.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>181</th>\n",
       "      <td>2020-11-07 00:00:00</td>\n",
       "      <td>Kolbotn</td>\n",
       "      <td>LSK Kvinner</td>\n",
       "      <td>Home</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2020.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>182</th>\n",
       "      <td>2020-07-04 00:00:00</td>\n",
       "      <td>Arna-Bjørnar</td>\n",
       "      <td>Røa</td>\n",
       "      <td>Away</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2020.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>183</th>\n",
       "      <td>2020-09-12 00:00:00</td>\n",
       "      <td>Sandviken</td>\n",
       "      <td>Lyn</td>\n",
       "      <td>Away</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2020.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                    Date          Team     Opponent Venue  y1: Team Score  \\\n",
       "174  2020-08-16 00:00:00       Kolbotn    Sandviken  Home             NaN   \n",
       "175  2020-12-06 00:00:00     Avaldsnes    Sandviken  Away             NaN   \n",
       "176  2020-10-03 00:00:00  Arna-Bjørnar  LSK Kvinner  Home             NaN   \n",
       "177  2020-07-04 00:00:00     Avaldsnes          Lyn  Home             NaN   \n",
       "178  2020-08-26 00:00:00  Arna-Bjørnar    Sandviken  Home             NaN   \n",
       "179  2020-09-27 00:00:00  Arna-Bjørnar    Sandviken  Away             NaN   \n",
       "180  2020-07-17 00:00:00           Røa  LSK Kvinner  Home             NaN   \n",
       "181  2020-11-07 00:00:00       Kolbotn  LSK Kvinner  Home             NaN   \n",
       "182  2020-07-04 00:00:00  Arna-Bjørnar          Røa  Away             NaN   \n",
       "183  2020-09-12 00:00:00     Sandviken          Lyn  Away             NaN   \n",
       "\n",
       "     y2: Opponent Score  Season  \n",
       "174                 NaN  2020.0  \n",
       "175                 NaN  2020.0  \n",
       "176                 NaN  2020.0  \n",
       "177                 NaN  2020.0  \n",
       "178                 NaN  2020.0  \n",
       "179                 NaN  2020.0  \n",
       "180                 NaN  2020.0  \n",
       "181                 NaN  2020.0  \n",
       "182                 NaN  2020.0  \n",
       "183                 NaN  2020.0  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Concatenate the 2020 data to the games data\n",
    "games_df = pd.concat([games_df, games_2020_df], axis=0)\n",
    "display(games_df.tail(10))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "0c8f8cf0",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Ensuring that Date feature is encoded with Datetime dtype for all rows\n",
    "games_df[\"Date\"] = pd.to_datetime(games_df[\"Date\"], yearfirst=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fe9b3fd6",
   "metadata": {},
   "source": [
    "[*back to top*](#top)\n",
    "### <a id=\"visualising-scores\"></a>3.4) Visualising the distribution of scores"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ba0a49e7",
   "metadata": {},
   "source": [
    "Before I proceed with the feature engineering, I want to explore some of the statistical properties of the exisiting features. I will start by visualising the distribution of the target variable, which will be the most important feature for training the machine learning models. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "cc4e4e35",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Separating the 2020 data from the games dataset so that it isn't present in the visualisation\n",
    "# I want to look at the test data as little as possible\n",
    "games_2020_df = games_df.loc[games_df[\"Season\"] == 2020].drop(columns=[\"y1: Team Score\", \"y2: Opponent Score\"])\n",
    "games_df = games_df.loc[games_df[\"Season\"] != 2020]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "4d12a102",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 936x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The mean number of goals is 1.6168341708542713\n"
     ]
    }
   ],
   "source": [
    "# Visualising the distribution of scores\n",
    "fig, ax = plt.subplots()\n",
    "n, bins, patches = ax.hist(games_df[\"y1: Team Score\"], label='Team Scores',\n",
    "                            stacked=False, color='darkorange', bins=25)\n",
    "\n",
    "# Adding a vertical line to distinguish the mean value\n",
    "ax.axvline(games_df[\"y1: Team Score\"].mean(), color='k', linestyle='dashed', linewidth=1, label=\"Mean Score\")\n",
    "ax.set_ylabel('Count', size=12)\n",
    "ax.set_xlabel('Number of Goals Scored', size=12)\n",
    "ax.set_title('Number of goals scored by a single team in Toppserien soccer matches, by count, 2017 - 2019', size=18)\n",
    "plt.legend(bbox_to_anchor=(0.97, 0.97))\n",
    "\n",
    "# Setting the plot size\n",
    "fig.set_figheight(8)\n",
    "fig.set_figwidth(13)\n",
    "\n",
    "# Displaying the plot\n",
    "plt.show()\n",
    "\n",
    "print(f\"The mean number of goals is {games_df['y1: Team Score'].mean()}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "79620ea7",
   "metadata": {},
   "source": [
    "As expected, we can see that most teams score between 0 and 2 goals in a single match. There are, however, a surprising amount of outliers. We see that many teams have scored between 4 and 6 goals, and there is even a team that managed to score 9 (!) goals in a single match. This has some important implications for how I am going to scale the data in the next part of the project.\n",
    "\n",
    "What's interesting to investigate further is whether the number of goals scored is related to whether the teams are playing a home game or not. I expect that teams generally perform slightly better on their home venue, since they'll have the extra morale from having more supporters cheering them on. Additionally, a better performance might simply be the cause of having to travel less to get to the match venue, and thus having more time to rest before the game. To determine whether there is any correlation between these two variables, I will visualise the same plot but with the scores aggregated by team and venue."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "86b4104c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1131.34x595.44 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Visualising the number of goals scored home and away \n",
    "# for each team over the course of the 2017, 2018 and 2019 seasons\n",
    "\n",
    "# Calculating the aggregated sum of goals for each team, for home and away.\n",
    "# Retrieving all matches played at home and calculate total goals for each team\n",
    "home_matches = games_df.loc[games_df[\"Venue\"] == \"Home\"]\n",
    "home_team_goals = home_matches.groupby(\"Team\")[\"y1: Team Score\"].sum()\n",
    "\n",
    "# Retrieving all matches played away and calculate total goals for each team\n",
    "away_matches = games_df.loc[games_df[\"Venue\"] == \"Away\"]\n",
    "away_team_goals = away_matches.groupby(\"Team\")[\"y1: Team Score\"].sum()\n",
    "\n",
    "# Concatenating the results\n",
    "team_goals = pd.concat([home_team_goals, away_team_goals], axis=1)\n",
    "team_goals.columns.values[[0, 1]] = ['Home goals', 'Away goals']\n",
    "team_goals = team_goals.reset_index()\n",
    "\n",
    "# Melting the results so that each team appears on two rows, once for home and once for away, with goals as value\n",
    "team_goals_melted = pd.melt(team_goals, id_vars=\"Team\", var_name=\"Venue\", value_name=\"Goals\")\n",
    "\n",
    "# Plotting the results\n",
    "g = sns.catplot(x='Team', y='Goals', hue='Venue', data=team_goals_melted, \n",
    "            kind='bar', height=8.27, aspect=19/10, legend=False)\n",
    "g.set_ylabels(\"Goals\", size=20)\n",
    "g.set_xlabels(\"Team\", size=20)\n",
    "g.set_yticklabels(size=14)\n",
    "g.fig.suptitle(\"Sum of scores for all teams in Toppserien, seasons 2017-2019\", size=18)\n",
    "plt.legend(loc='upper right', fontsize=18)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4e50757d",
   "metadata": {},
   "source": [
    "The plot clearly shows that playing at home is a significant advantage. There are only two teams that haven't scored more goals at home than away, and one of them (*IF Fløya*) is only present in a few samples in the training data. We can also see that the team *LSK Kvinner* has been dominating the championships for the last few years, followed by a handful of other teams. I therefore think that a dummy variable indicating whether the team (or opponent) finished in the top 5 in last year's season might be a very good indicator for how many goals each team is going to score."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "327c21a4",
   "metadata": {},
   "source": [
    "### <a id=\"game-season\"></a>3.5) Adding game season and game number in season"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1eaba55e",
   "metadata": {},
   "source": [
    "I will now derive some new, simple features from the games data while also adding some of the features from the team and tables datasets. I hope that these features might provide what is needed for a solid baseline for predicting the outcome of the 2020 season. More advanced feature engineering will likely occur in the modelling stage. \n",
    "\n",
    "Firstly, I would like to denote which game in the season the teams are playing. Since the Toppserien championship consists of 264 games every season such that all 12 team plays against every other team both home and away, I believe that less experienced teams tend to experience more \"fatigue\" over the course of the season than the more experienced ones. Some teams are also better than others at avoiding injuries, as well as penalty cards. Teams that have a high number of penalty cards may be likely to experience matches later in the season where one or more of their important players aren't allowed on the field, perhaps because they recieved one too many yellow cards over the past few games. In this way, \"aggressive\" teams - and teams that don't take proper care of their players - might perform progressively worse over the course of the season. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "c5e409ac",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Re-joining the games data\n",
    "games_df = pd.concat([games_df, games_2020_df], axis=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "ece107b1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Date</th>\n",
       "      <th>Team</th>\n",
       "      <th>Opponent</th>\n",
       "      <th>Venue</th>\n",
       "      <th>y1: Team Score</th>\n",
       "      <th>y2: Opponent Score</th>\n",
       "      <th>Season</th>\n",
       "      <th>Game # in season</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>Medkila</td>\n",
       "      <td>Sandviken</td>\n",
       "      <td>Home</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2017</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>Arna-Bjørnar</td>\n",
       "      <td>Grand Bodø</td>\n",
       "      <td>Away</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2017</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>Stabæk</td>\n",
       "      <td>Røa</td>\n",
       "      <td>Away</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2017</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>Klepp</td>\n",
       "      <td>LSK Kvinner</td>\n",
       "      <td>Away</td>\n",
       "      <td>1.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>2017</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>Kolbotn</td>\n",
       "      <td>Trondheims-Ørn</td>\n",
       "      <td>Away</td>\n",
       "      <td>2.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2017</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        Date          Team        Opponent Venue  y1: Team Score  \\\n",
       "0 2017-04-17       Medkila       Sandviken  Home             1.0   \n",
       "1 2017-04-17  Arna-Bjørnar      Grand Bodø  Away             2.0   \n",
       "2 2017-04-17        Stabæk             Røa  Away             2.0   \n",
       "3 2017-04-17         Klepp     LSK Kvinner  Away             1.0   \n",
       "4 2017-04-17       Kolbotn  Trondheims-Ørn  Away             2.0   \n",
       "\n",
       "   y2: Opponent Score  Season  Game # in season  \n",
       "0                 1.0    2017               1.0  \n",
       "1                 2.0    2017               1.0  \n",
       "2                 1.0    2017               2.0  \n",
       "3                 6.0    2017               1.0  \n",
       "4                 3.0    2017               1.0  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Date</th>\n",
       "      <th>Team</th>\n",
       "      <th>Opponent</th>\n",
       "      <th>Venue</th>\n",
       "      <th>y1: Team Score</th>\n",
       "      <th>y2: Opponent Score</th>\n",
       "      <th>Season</th>\n",
       "      <th>Game # in season</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>179</th>\n",
       "      <td>2020-09-27</td>\n",
       "      <td>Arna-Bjørnar</td>\n",
       "      <td>Sandviken</td>\n",
       "      <td>Away</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2020</td>\n",
       "      <td>17.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>180</th>\n",
       "      <td>2020-07-17</td>\n",
       "      <td>Røa</td>\n",
       "      <td>LSK Kvinner</td>\n",
       "      <td>Home</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2020</td>\n",
       "      <td>18.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>181</th>\n",
       "      <td>2020-11-07</td>\n",
       "      <td>Kolbotn</td>\n",
       "      <td>LSK Kvinner</td>\n",
       "      <td>Home</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2020</td>\n",
       "      <td>20.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>182</th>\n",
       "      <td>2020-07-04</td>\n",
       "      <td>Arna-Bjørnar</td>\n",
       "      <td>Røa</td>\n",
       "      <td>Away</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2020</td>\n",
       "      <td>18.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>183</th>\n",
       "      <td>2020-09-12</td>\n",
       "      <td>Sandviken</td>\n",
       "      <td>Lyn</td>\n",
       "      <td>Away</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2020</td>\n",
       "      <td>18.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          Date          Team     Opponent Venue  y1: Team Score  \\\n",
       "179 2020-09-27  Arna-Bjørnar    Sandviken  Away             NaN   \n",
       "180 2020-07-17           Røa  LSK Kvinner  Home             NaN   \n",
       "181 2020-11-07       Kolbotn  LSK Kvinner  Home             NaN   \n",
       "182 2020-07-04  Arna-Bjørnar          Røa  Away             NaN   \n",
       "183 2020-09-12     Sandviken          Lyn  Away             NaN   \n",
       "\n",
       "     y2: Opponent Score  Season  Game # in season  \n",
       "179                 NaN    2020              17.0  \n",
       "180                 NaN    2020              18.0  \n",
       "181                 NaN    2020              20.0  \n",
       "182                 NaN    2020              18.0  \n",
       "183                 NaN    2020              18.0  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Creating a new feature with only the year of the championship season\n",
    "games_df[\"Season\"] = games_df[\"Date\"].dt.year\n",
    "\n",
    "# Creating a new feature with the count of games the team has played, inclusive of current game\n",
    "teams = set(games_df[\"Team\"].values)\n",
    "games_played = {team: 0 for team in teams}\n",
    "current_season = \"2017\"\n",
    "\n",
    "# Looping over the dataframe, calculating # of matches played and inserting it to new 'Game # in season' feature\n",
    "for index, row in games_df.iterrows(): \n",
    "    if row[\"Season\"] != current_season:\n",
    "        current_season = row[\"Season\"]\n",
    "        games_played = dict.fromkeys(games_played, 0)  # resetting the game count\n",
    "    games_played[row[\"Team\"]] += 1\n",
    "    games_df.loc[[index], [\"Game # in season\"]] = games_played[row[\"Team\"]]\n",
    "\n",
    "# Displaying the result\n",
    "display(games_df.head())\n",
    "display(games_df.tail())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "40b2c398",
   "metadata": {},
   "source": [
    "### <a id=\"ahsd\"></a>3.6) Adding the average historical score difference"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f6cb2883",
   "metadata": {},
   "source": [
    "Another interesting feature I would like to calculate is the average historical score difference (let's call it AHSD) between the two teams playing each other. If Stabæk tends to beat Røa by a significant lead every time they play against each other, the historical goal difference should be represented explicitly as a new variable. I would like to experiment more with this metric later, perhaps by using a weighted moving average so that more recent games are weighted more. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "13241dd7",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Season</th>\n",
       "      <th>Team</th>\n",
       "      <th>Opponent</th>\n",
       "      <th>Venue</th>\n",
       "      <th>Game # in season</th>\n",
       "      <th>Avg. Historical Score Diff</th>\n",
       "      <th>Date</th>\n",
       "      <th>y1: Team Score</th>\n",
       "      <th>y2: Opponent Score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>179</th>\n",
       "      <td>2020</td>\n",
       "      <td>Arna-Bjørnar</td>\n",
       "      <td>Sandviken</td>\n",
       "      <td>Away</td>\n",
       "      <td>17.0</td>\n",
       "      <td>-0.571429</td>\n",
       "      <td>2020-09-27</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>180</th>\n",
       "      <td>2020</td>\n",
       "      <td>Røa</td>\n",
       "      <td>LSK Kvinner</td>\n",
       "      <td>Home</td>\n",
       "      <td>18.0</td>\n",
       "      <td>-2.000000</td>\n",
       "      <td>2020-07-17</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>181</th>\n",
       "      <td>2020</td>\n",
       "      <td>Kolbotn</td>\n",
       "      <td>LSK Kvinner</td>\n",
       "      <td>Home</td>\n",
       "      <td>20.0</td>\n",
       "      <td>-1.714286</td>\n",
       "      <td>2020-11-07</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>182</th>\n",
       "      <td>2020</td>\n",
       "      <td>Arna-Bjørnar</td>\n",
       "      <td>Røa</td>\n",
       "      <td>Away</td>\n",
       "      <td>18.0</td>\n",
       "      <td>-0.571429</td>\n",
       "      <td>2020-07-04</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>183</th>\n",
       "      <td>2020</td>\n",
       "      <td>Sandviken</td>\n",
       "      <td>Lyn</td>\n",
       "      <td>Away</td>\n",
       "      <td>18.0</td>\n",
       "      <td>1.800000</td>\n",
       "      <td>2020-09-12</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     Season          Team     Opponent Venue  Game # in season  \\\n",
       "179    2020  Arna-Bjørnar    Sandviken  Away              17.0   \n",
       "180    2020           Røa  LSK Kvinner  Home              18.0   \n",
       "181    2020       Kolbotn  LSK Kvinner  Home              20.0   \n",
       "182    2020  Arna-Bjørnar          Røa  Away              18.0   \n",
       "183    2020     Sandviken          Lyn  Away              18.0   \n",
       "\n",
       "     Avg. Historical Score Diff       Date  y1: Team Score  y2: Opponent Score  \n",
       "179                   -0.571429 2020-09-27             NaN                 NaN  \n",
       "180                   -2.000000 2020-07-17             NaN                 NaN  \n",
       "181                   -1.714286 2020-11-07             NaN                 NaN  \n",
       "182                   -0.571429 2020-07-04             NaN                 NaN  \n",
       "183                    1.800000 2020-09-12             NaN                 NaN  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Calculating the average historical score differance (AHSD) between the two teams playing\n",
    "score_diffs = {playing_team: {team : [0, 0] for team in teams if team != playing_team} for playing_team in teams}\n",
    "\n",
    "for index, row in games_df.iterrows():\n",
    "    # Insert the value for the AHSD\n",
    "    try:\n",
    "        ahsd_team_vs_opponent = score_diffs[row['Team']][row['Opponent']][0] / score_diffs[row['Team']][row['Opponent']][1]\n",
    "        games_df.loc[[index], [\"Avg. Historical Score Diff\"]] = ahsd_team_vs_opponent\n",
    "    except ZeroDivisionError:\n",
    "        # This error implicates that the two teams have never played against each other before\n",
    "        games_df.loc[[index], [\"Avg. Historical Score Diff\"]] = 0\n",
    "    \n",
    "    # update the AHSD after the current game, unless we are calculating AHSD for games in 2020\n",
    "    if (not pd.isnull(row[\"y1: Team Score\"])) and (not pd.isnull(row[\"y2: Opponent Score\"])):\n",
    "        score_diffs[row[\"Team\"]][row[\"Opponent\"]][0] += (row[\"y1: Team Score\"] - row[\"y2: Opponent Score\"])\n",
    "\n",
    "    # update count of how many times the team has played against the opponent so far\n",
    "    score_diffs[row[\"Team\"]][row[\"Opponent\"]][1] += 1\n",
    "    \n",
    "    \n",
    "# Reordering the columns so that the predictor variable appears at the end\n",
    "games_df = games_df[[\"Season\", \"Team\", \"Opponent\", \"Venue\", \"Game # in season\", \"Avg. Historical Score Diff\", \"Date\",  \"y1: Team Score\", \"y2: Opponent Score\"]]\n",
    "\n",
    "display(games_df.tail())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cf94fd21",
   "metadata": {},
   "source": [
    "### <a id=\"avg-nr-goals\"></a>3.7) Adding the average number of goals per 90 min last season"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "788b01ee",
   "metadata": {},
   "source": [
    "Next, I will load the teams data and retrieve the avg. number of goals the team scored the previous season per 90 min. This value will be inserted into the games dataframe, along with the same metric for the opposing teams. The reason I am basing this metric on the previous season is that I won't be able to calculate a moving average for the 2020 season since the model will predict the outcome of all games in the season at the same time. The model will only have access to the moving average for the 2019 season, and since ~4 months pass between the 2019 and 2020 seasons any \"streaks\" or other indicators of a team's current \"form\" will be rather useless. The avg. number of goals scored should however indicate whether a team is aggressive or defensive, and along with features for received goals the model should be able to predict the end results for games with high- and low-scoring teams more accurately."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "molecular-landing",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead tr th {\n",
       "        text-align: left;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>Unnamed: 0_level_0</th>\n",
       "      <th>Unnamed: 1_level_0</th>\n",
       "      <th>Unnamed: 2_level_0</th>\n",
       "      <th>Unnamed: 3_level_0</th>\n",
       "      <th colspan=\"4\" halign=\"left\">Playing Time</th>\n",
       "      <th colspan=\"7\" halign=\"left\">Performance</th>\n",
       "      <th colspan=\"5\" halign=\"left\">Per 90 Minutes</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>Squad</th>\n",
       "      <th># Pl</th>\n",
       "      <th>Age</th>\n",
       "      <th>Poss</th>\n",
       "      <th>MP</th>\n",
       "      <th>Starts</th>\n",
       "      <th>Min</th>\n",
       "      <th>90s</th>\n",
       "      <th>Gls</th>\n",
       "      <th>Ast</th>\n",
       "      <th>G-PK</th>\n",
       "      <th>PK</th>\n",
       "      <th>PKatt</th>\n",
       "      <th>CrdY</th>\n",
       "      <th>CrdR</th>\n",
       "      <th>Gls</th>\n",
       "      <th>Ast</th>\n",
       "      <th>G+A</th>\n",
       "      <th>G-PK</th>\n",
       "      <th>G+A-PK</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Arna-Bjørnar</td>\n",
       "      <td>22</td>\n",
       "      <td>22.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>22</td>\n",
       "      <td>242</td>\n",
       "      <td>1980</td>\n",
       "      <td>22.0</td>\n",
       "      <td>37</td>\n",
       "      <td>NaN</td>\n",
       "      <td>36</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>20</td>\n",
       "      <td>1</td>\n",
       "      <td>1.68</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.64</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Avaldsnes</td>\n",
       "      <td>23</td>\n",
       "      <td>25.5</td>\n",
       "      <td>NaN</td>\n",
       "      <td>22</td>\n",
       "      <td>242</td>\n",
       "      <td>1980</td>\n",
       "      <td>22.0</td>\n",
       "      <td>49</td>\n",
       "      <td>NaN</td>\n",
       "      <td>48</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>22</td>\n",
       "      <td>1</td>\n",
       "      <td>2.23</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.18</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Grand Bodø</td>\n",
       "      <td>22</td>\n",
       "      <td>22.7</td>\n",
       "      <td>NaN</td>\n",
       "      <td>22</td>\n",
       "      <td>242</td>\n",
       "      <td>1980</td>\n",
       "      <td>22.0</td>\n",
       "      <td>30</td>\n",
       "      <td>NaN</td>\n",
       "      <td>29</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>19</td>\n",
       "      <td>0</td>\n",
       "      <td>1.36</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.32</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Klepp</td>\n",
       "      <td>19</td>\n",
       "      <td>23.5</td>\n",
       "      <td>NaN</td>\n",
       "      <td>22</td>\n",
       "      <td>242</td>\n",
       "      <td>1980</td>\n",
       "      <td>22.0</td>\n",
       "      <td>33</td>\n",
       "      <td>NaN</td>\n",
       "      <td>31</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>13</td>\n",
       "      <td>2</td>\n",
       "      <td>1.50</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.41</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Kolbotn</td>\n",
       "      <td>21</td>\n",
       "      <td>21.9</td>\n",
       "      <td>NaN</td>\n",
       "      <td>22</td>\n",
       "      <td>242</td>\n",
       "      <td>1980</td>\n",
       "      <td>22.0</td>\n",
       "      <td>16</td>\n",
       "      <td>NaN</td>\n",
       "      <td>16</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>8</td>\n",
       "      <td>0</td>\n",
       "      <td>0.73</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.73</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>LSK Kvinner</td>\n",
       "      <td>18</td>\n",
       "      <td>24.2</td>\n",
       "      <td>NaN</td>\n",
       "      <td>22</td>\n",
       "      <td>242</td>\n",
       "      <td>1980</td>\n",
       "      <td>22.0</td>\n",
       "      <td>67</td>\n",
       "      <td>NaN</td>\n",
       "      <td>65</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>11</td>\n",
       "      <td>0</td>\n",
       "      <td>3.05</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.95</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Medkila</td>\n",
       "      <td>26</td>\n",
       "      <td>20.9</td>\n",
       "      <td>NaN</td>\n",
       "      <td>22</td>\n",
       "      <td>242</td>\n",
       "      <td>1980</td>\n",
       "      <td>22.0</td>\n",
       "      <td>9</td>\n",
       "      <td>NaN</td>\n",
       "      <td>9</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>13</td>\n",
       "      <td>2</td>\n",
       "      <td>0.41</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.41</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Røa</td>\n",
       "      <td>22</td>\n",
       "      <td>22.8</td>\n",
       "      <td>NaN</td>\n",
       "      <td>22</td>\n",
       "      <td>242</td>\n",
       "      <td>1980</td>\n",
       "      <td>22.0</td>\n",
       "      <td>42</td>\n",
       "      <td>NaN</td>\n",
       "      <td>42</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>22</td>\n",
       "      <td>1</td>\n",
       "      <td>1.91</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.91</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>Sandviken</td>\n",
       "      <td>22</td>\n",
       "      <td>24.4</td>\n",
       "      <td>NaN</td>\n",
       "      <td>22</td>\n",
       "      <td>242</td>\n",
       "      <td>1980</td>\n",
       "      <td>22.0</td>\n",
       "      <td>23</td>\n",
       "      <td>NaN</td>\n",
       "      <td>23</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>10</td>\n",
       "      <td>0</td>\n",
       "      <td>1.05</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.05</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>Stabæk</td>\n",
       "      <td>20</td>\n",
       "      <td>25.5</td>\n",
       "      <td>NaN</td>\n",
       "      <td>22</td>\n",
       "      <td>242</td>\n",
       "      <td>1980</td>\n",
       "      <td>22.0</td>\n",
       "      <td>44</td>\n",
       "      <td>NaN</td>\n",
       "      <td>44</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>2.00</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.00</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>Trondheims-Ørn</td>\n",
       "      <td>22</td>\n",
       "      <td>22.1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>22</td>\n",
       "      <td>242</td>\n",
       "      <td>1980</td>\n",
       "      <td>22.0</td>\n",
       "      <td>36</td>\n",
       "      <td>NaN</td>\n",
       "      <td>36</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>9</td>\n",
       "      <td>0</td>\n",
       "      <td>1.64</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.64</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>Vålerenga</td>\n",
       "      <td>23</td>\n",
       "      <td>25.8</td>\n",
       "      <td>NaN</td>\n",
       "      <td>22</td>\n",
       "      <td>242</td>\n",
       "      <td>1980</td>\n",
       "      <td>22.0</td>\n",
       "      <td>38</td>\n",
       "      <td>NaN</td>\n",
       "      <td>38</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>16</td>\n",
       "      <td>2</td>\n",
       "      <td>1.73</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.73</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Unnamed: 0_level_0 Unnamed: 1_level_0 Unnamed: 2_level_0  \\\n",
       "                Squad               # Pl                Age   \n",
       "0        Arna-Bjørnar                 22               22.0   \n",
       "1           Avaldsnes                 23               25.5   \n",
       "2          Grand Bodø                 22               22.7   \n",
       "3               Klepp                 19               23.5   \n",
       "4             Kolbotn                 21               21.9   \n",
       "5         LSK Kvinner                 18               24.2   \n",
       "6             Medkila                 26               20.9   \n",
       "7                 Røa                 22               22.8   \n",
       "8           Sandviken                 22               24.4   \n",
       "9              Stabæk                 20               25.5   \n",
       "10     Trondheims-Ørn                 22               22.1   \n",
       "11          Vålerenga                 23               25.8   \n",
       "\n",
       "   Unnamed: 3_level_0 Playing Time                    Performance              \\\n",
       "                 Poss           MP Starts   Min   90s         Gls Ast G-PK PK   \n",
       "0                 NaN           22    242  1980  22.0          37 NaN   36  1   \n",
       "1                 NaN           22    242  1980  22.0          49 NaN   48  1   \n",
       "2                 NaN           22    242  1980  22.0          30 NaN   29  1   \n",
       "3                 NaN           22    242  1980  22.0          33 NaN   31  2   \n",
       "4                 NaN           22    242  1980  22.0          16 NaN   16  0   \n",
       "5                 NaN           22    242  1980  22.0          67 NaN   65  2   \n",
       "6                 NaN           22    242  1980  22.0           9 NaN    9  0   \n",
       "7                 NaN           22    242  1980  22.0          42 NaN   42  0   \n",
       "8                 NaN           22    242  1980  22.0          23 NaN   23  0   \n",
       "9                 NaN           22    242  1980  22.0          44 NaN   44  0   \n",
       "10                NaN           22    242  1980  22.0          36 NaN   36  0   \n",
       "11                NaN           22    242  1980  22.0          38 NaN   38  0   \n",
       "\n",
       "                   Per 90 Minutes                       \n",
       "   PKatt CrdY CrdR            Gls Ast G+A  G-PK G+A-PK  \n",
       "0      1   20    1           1.68 NaN NaN  1.64    NaN  \n",
       "1      1   22    1           2.23 NaN NaN  2.18    NaN  \n",
       "2      1   19    0           1.36 NaN NaN  1.32    NaN  \n",
       "3      2   13    2           1.50 NaN NaN  1.41    NaN  \n",
       "4      0    8    0           0.73 NaN NaN  0.73    NaN  \n",
       "5      2   11    0           3.05 NaN NaN  2.95    NaN  \n",
       "6      0   13    2           0.41 NaN NaN  0.41    NaN  \n",
       "7      0   22    1           1.91 NaN NaN  1.91    NaN  \n",
       "8      0   10    0           1.05 NaN NaN  1.05    NaN  \n",
       "9      0    4    0           2.00 NaN NaN  2.00    NaN  \n",
       "10     0    9    0           1.64 NaN NaN  1.64    NaN  \n",
       "11     0   16    2           1.73 NaN NaN  1.73    NaN  "
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Loading the team-stats dataset\n",
    "teams_2017_df = pd.read_html(\"Raw Data/2017/team-stats.xls\", encoding=\"UTF-8\")[0]\n",
    "teams_2018_df = pd.read_html(\"Raw Data/2018/team-stats.xls\", encoding=\"UTF-8\")[0]\n",
    "teams_2019_df = pd.read_html(\"Raw Data/2019/team-stats.xls\", encoding=\"UTF-8\")[0]\n",
    "\n",
    "teams_2017_df"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "38767e32",
   "metadata": {},
   "source": [
    "Before I continue, I will quickly remove features that only contain missing values from the teams dataset. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "f72b228f",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead tr th {\n",
       "        text-align: left;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>Unnamed: 0_level_0</th>\n",
       "      <th>Unnamed: 1_level_0</th>\n",
       "      <th>Unnamed: 2_level_0</th>\n",
       "      <th colspan=\"4\" halign=\"left\">Playing Time</th>\n",
       "      <th colspan=\"6\" halign=\"left\">Performance</th>\n",
       "      <th colspan=\"2\" halign=\"left\">Per 90 Minutes</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th>Squad</th>\n",
       "      <th># Pl</th>\n",
       "      <th>Age</th>\n",
       "      <th>MP</th>\n",
       "      <th>Starts</th>\n",
       "      <th>Min</th>\n",
       "      <th>90s</th>\n",
       "      <th>Gls</th>\n",
       "      <th>G-PK</th>\n",
       "      <th>PK</th>\n",
       "      <th>PKatt</th>\n",
       "      <th>CrdY</th>\n",
       "      <th>CrdR</th>\n",
       "      <th>Gls</th>\n",
       "      <th>G-PK</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Arna-Bjørnar</td>\n",
       "      <td>22</td>\n",
       "      <td>22.0</td>\n",
       "      <td>22</td>\n",
       "      <td>242</td>\n",
       "      <td>1980</td>\n",
       "      <td>22.0</td>\n",
       "      <td>37</td>\n",
       "      <td>36</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>20</td>\n",
       "      <td>1</td>\n",
       "      <td>1.68</td>\n",
       "      <td>1.64</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Avaldsnes</td>\n",
       "      <td>23</td>\n",
       "      <td>25.5</td>\n",
       "      <td>22</td>\n",
       "      <td>242</td>\n",
       "      <td>1980</td>\n",
       "      <td>22.0</td>\n",
       "      <td>49</td>\n",
       "      <td>48</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>22</td>\n",
       "      <td>1</td>\n",
       "      <td>2.23</td>\n",
       "      <td>2.18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Grand Bodø</td>\n",
       "      <td>22</td>\n",
       "      <td>22.7</td>\n",
       "      <td>22</td>\n",
       "      <td>242</td>\n",
       "      <td>1980</td>\n",
       "      <td>22.0</td>\n",
       "      <td>30</td>\n",
       "      <td>29</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>19</td>\n",
       "      <td>0</td>\n",
       "      <td>1.36</td>\n",
       "      <td>1.32</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Klepp</td>\n",
       "      <td>19</td>\n",
       "      <td>23.5</td>\n",
       "      <td>22</td>\n",
       "      <td>242</td>\n",
       "      <td>1980</td>\n",
       "      <td>22.0</td>\n",
       "      <td>33</td>\n",
       "      <td>31</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>13</td>\n",
       "      <td>2</td>\n",
       "      <td>1.50</td>\n",
       "      <td>1.41</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Kolbotn</td>\n",
       "      <td>21</td>\n",
       "      <td>21.9</td>\n",
       "      <td>22</td>\n",
       "      <td>242</td>\n",
       "      <td>1980</td>\n",
       "      <td>22.0</td>\n",
       "      <td>16</td>\n",
       "      <td>16</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>8</td>\n",
       "      <td>0</td>\n",
       "      <td>0.73</td>\n",
       "      <td>0.73</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>LSK Kvinner</td>\n",
       "      <td>18</td>\n",
       "      <td>24.2</td>\n",
       "      <td>22</td>\n",
       "      <td>242</td>\n",
       "      <td>1980</td>\n",
       "      <td>22.0</td>\n",
       "      <td>67</td>\n",
       "      <td>65</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>11</td>\n",
       "      <td>0</td>\n",
       "      <td>3.05</td>\n",
       "      <td>2.95</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Medkila</td>\n",
       "      <td>26</td>\n",
       "      <td>20.9</td>\n",
       "      <td>22</td>\n",
       "      <td>242</td>\n",
       "      <td>1980</td>\n",
       "      <td>22.0</td>\n",
       "      <td>9</td>\n",
       "      <td>9</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>13</td>\n",
       "      <td>2</td>\n",
       "      <td>0.41</td>\n",
       "      <td>0.41</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Røa</td>\n",
       "      <td>22</td>\n",
       "      <td>22.8</td>\n",
       "      <td>22</td>\n",
       "      <td>242</td>\n",
       "      <td>1980</td>\n",
       "      <td>22.0</td>\n",
       "      <td>42</td>\n",
       "      <td>42</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>22</td>\n",
       "      <td>1</td>\n",
       "      <td>1.91</td>\n",
       "      <td>1.91</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>Sandviken</td>\n",
       "      <td>22</td>\n",
       "      <td>24.4</td>\n",
       "      <td>22</td>\n",
       "      <td>242</td>\n",
       "      <td>1980</td>\n",
       "      <td>22.0</td>\n",
       "      <td>23</td>\n",
       "      <td>23</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>10</td>\n",
       "      <td>0</td>\n",
       "      <td>1.05</td>\n",
       "      <td>1.05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>Stabæk</td>\n",
       "      <td>20</td>\n",
       "      <td>25.5</td>\n",
       "      <td>22</td>\n",
       "      <td>242</td>\n",
       "      <td>1980</td>\n",
       "      <td>22.0</td>\n",
       "      <td>44</td>\n",
       "      <td>44</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>2.00</td>\n",
       "      <td>2.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>Trondheims-Ørn</td>\n",
       "      <td>22</td>\n",
       "      <td>22.1</td>\n",
       "      <td>22</td>\n",
       "      <td>242</td>\n",
       "      <td>1980</td>\n",
       "      <td>22.0</td>\n",
       "      <td>36</td>\n",
       "      <td>36</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>9</td>\n",
       "      <td>0</td>\n",
       "      <td>1.64</td>\n",
       "      <td>1.64</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>Vålerenga</td>\n",
       "      <td>23</td>\n",
       "      <td>25.8</td>\n",
       "      <td>22</td>\n",
       "      <td>242</td>\n",
       "      <td>1980</td>\n",
       "      <td>22.0</td>\n",
       "      <td>38</td>\n",
       "      <td>38</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>16</td>\n",
       "      <td>2</td>\n",
       "      <td>1.73</td>\n",
       "      <td>1.73</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Unnamed: 0_level_0 Unnamed: 1_level_0 Unnamed: 2_level_0 Playing Time  \\\n",
       "                Squad               # Pl                Age           MP   \n",
       "0        Arna-Bjørnar                 22               22.0           22   \n",
       "1           Avaldsnes                 23               25.5           22   \n",
       "2          Grand Bodø                 22               22.7           22   \n",
       "3               Klepp                 19               23.5           22   \n",
       "4             Kolbotn                 21               21.9           22   \n",
       "5         LSK Kvinner                 18               24.2           22   \n",
       "6             Medkila                 26               20.9           22   \n",
       "7                 Røa                 22               22.8           22   \n",
       "8           Sandviken                 22               24.4           22   \n",
       "9              Stabæk                 20               25.5           22   \n",
       "10     Trondheims-Ørn                 22               22.1           22   \n",
       "11          Vålerenga                 23               25.8           22   \n",
       "\n",
       "                      Performance                         Per 90 Minutes        \n",
       "   Starts   Min   90s         Gls G-PK PK PKatt CrdY CrdR            Gls  G-PK  \n",
       "0     242  1980  22.0          37   36  1     1   20    1           1.68  1.64  \n",
       "1     242  1980  22.0          49   48  1     1   22    1           2.23  2.18  \n",
       "2     242  1980  22.0          30   29  1     1   19    0           1.36  1.32  \n",
       "3     242  1980  22.0          33   31  2     2   13    2           1.50  1.41  \n",
       "4     242  1980  22.0          16   16  0     0    8    0           0.73  0.73  \n",
       "5     242  1980  22.0          67   65  2     2   11    0           3.05  2.95  \n",
       "6     242  1980  22.0           9    9  0     0   13    2           0.41  0.41  \n",
       "7     242  1980  22.0          42   42  0     0   22    1           1.91  1.91  \n",
       "8     242  1980  22.0          23   23  0     0   10    0           1.05  1.05  \n",
       "9     242  1980  22.0          44   44  0     0    4    0           2.00  2.00  \n",
       "10    242  1980  22.0          36   36  0     0    9    0           1.64  1.64  \n",
       "11    242  1980  22.0          38   38  0     0   16    2           1.73  1.73  "
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Remove features with only NaN values\n",
    "for df in [teams_2017_df, teams_2018_df, teams_2019_df]:\n",
    "    df.dropna(how=\"all\", axis=1, inplace=True)\n",
    "    \n",
    "teams_2017_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "092c9c4e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1.6401515151515151\n",
      "Half a standard deviation from the mean of goals scored for a single team in the 2017 season is 0.8301674185884825\n",
      "Half a standard deviation from the mean of goals scored for single team in the 2018 season is 0.9257405057019253\n",
      "Half a standard deviation from the mean of goals scored for a single team in the 2019 season is 0.8286568783754502\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Season</th>\n",
       "      <th>Team</th>\n",
       "      <th>Opponent</th>\n",
       "      <th>Venue</th>\n",
       "      <th>Game # in season</th>\n",
       "      <th>Avg. Historical Score Diff</th>\n",
       "      <th>Date</th>\n",
       "      <th>Team's goals per 90 min last season</th>\n",
       "      <th>Opponent's goals per 90 min last season</th>\n",
       "      <th>y1: Team Score</th>\n",
       "      <th>y2: Opponent Score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2017</td>\n",
       "      <td>Medkila</td>\n",
       "      <td>Sandviken</td>\n",
       "      <td>Home</td>\n",
       "      <td>1.0</td>\n",
       "      <td>-0.166667</td>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>1.640152</td>\n",
       "      <td>1.640152</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2017</td>\n",
       "      <td>Arna-Bjørnar</td>\n",
       "      <td>Grand Bodø</td>\n",
       "      <td>Away</td>\n",
       "      <td>1.0</td>\n",
       "      <td>-0.166667</td>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>1.640152</td>\n",
       "      <td>1.640152</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2017</td>\n",
       "      <td>Stabæk</td>\n",
       "      <td>Røa</td>\n",
       "      <td>Away</td>\n",
       "      <td>2.0</td>\n",
       "      <td>-1.833333</td>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>1.640152</td>\n",
       "      <td>1.640152</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2017</td>\n",
       "      <td>Klepp</td>\n",
       "      <td>LSK Kvinner</td>\n",
       "      <td>Away</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.250000</td>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>1.640152</td>\n",
       "      <td>1.640152</td>\n",
       "      <td>1.0</td>\n",
       "      <td>6.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2017</td>\n",
       "      <td>Kolbotn</td>\n",
       "      <td>Trondheims-Ørn</td>\n",
       "      <td>Away</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.166667</td>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>1.640152</td>\n",
       "      <td>1.640152</td>\n",
       "      <td>2.0</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>2017</td>\n",
       "      <td>Vålerenga</td>\n",
       "      <td>Avaldsnes</td>\n",
       "      <td>Away</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>1.640152</td>\n",
       "      <td>1.640152</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>2017</td>\n",
       "      <td>Sandviken</td>\n",
       "      <td>Medkila</td>\n",
       "      <td>Away</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>1.640152</td>\n",
       "      <td>1.640152</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>2017</td>\n",
       "      <td>Trondheims-Ørn</td>\n",
       "      <td>Kolbotn</td>\n",
       "      <td>Home</td>\n",
       "      <td>1.0</td>\n",
       "      <td>-0.833333</td>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>1.640152</td>\n",
       "      <td>1.640152</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>2017</td>\n",
       "      <td>LSK Kvinner</td>\n",
       "      <td>Klepp</td>\n",
       "      <td>Home</td>\n",
       "      <td>2.0</td>\n",
       "      <td>-0.142857</td>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>1.640152</td>\n",
       "      <td>1.640152</td>\n",
       "      <td>6.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>2017</td>\n",
       "      <td>Røa</td>\n",
       "      <td>Stabæk</td>\n",
       "      <td>Home</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>1.640152</td>\n",
       "      <td>1.640152</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Season            Team        Opponent Venue  Game # in season  \\\n",
       "0    2017         Medkila       Sandviken  Home               1.0   \n",
       "1    2017    Arna-Bjørnar      Grand Bodø  Away               1.0   \n",
       "2    2017          Stabæk             Røa  Away               2.0   \n",
       "3    2017           Klepp     LSK Kvinner  Away               1.0   \n",
       "4    2017         Kolbotn  Trondheims-Ørn  Away               1.0   \n",
       "5    2017       Vålerenga       Avaldsnes  Away               1.0   \n",
       "6    2017       Sandviken         Medkila  Away               1.0   \n",
       "7    2017  Trondheims-Ørn         Kolbotn  Home               1.0   \n",
       "8    2017     LSK Kvinner           Klepp  Home               2.0   \n",
       "9    2017             Røa          Stabæk  Home               2.0   \n",
       "\n",
       "   Avg. Historical Score Diff       Date  Team's goals per 90 min last season  \\\n",
       "0                   -0.166667 2017-04-17                             1.640152   \n",
       "1                   -0.166667 2017-04-17                             1.640152   \n",
       "2                   -1.833333 2017-04-17                             1.640152   \n",
       "3                    1.250000 2017-04-17                             1.640152   \n",
       "4                    1.166667 2017-04-17                             1.640152   \n",
       "5                    0.000000 2017-04-17                             1.640152   \n",
       "6                    0.666667 2017-04-17                             1.640152   \n",
       "7                   -0.833333 2017-04-17                             1.640152   \n",
       "8                   -0.142857 2017-04-17                             1.640152   \n",
       "9                    0.000000 2017-04-17                             1.640152   \n",
       "\n",
       "   Opponent's goals per 90 min last season  y1: Team Score  y2: Opponent Score  \n",
       "0                                 1.640152             1.0                 1.0  \n",
       "1                                 1.640152             2.0                 2.0  \n",
       "2                                 1.640152             2.0                 1.0  \n",
       "3                                 1.640152             1.0                 6.0  \n",
       "4                                 1.640152             2.0                 3.0  \n",
       "5                                 1.640152             1.0                 2.0  \n",
       "6                                 1.640152             1.0                 1.0  \n",
       "7                                 1.640152             3.0                 2.0  \n",
       "8                                 1.640152             6.0                 1.0  \n",
       "9                                 1.640152             1.0                 2.0  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Season</th>\n",
       "      <th>Team</th>\n",
       "      <th>Opponent</th>\n",
       "      <th>Venue</th>\n",
       "      <th>Game # in season</th>\n",
       "      <th>Avg. Historical Score Diff</th>\n",
       "      <th>Date</th>\n",
       "      <th>Team's goals per 90 min last season</th>\n",
       "      <th>Opponent's goals per 90 min last season</th>\n",
       "      <th>y1: Team Score</th>\n",
       "      <th>y2: Opponent Score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>174</th>\n",
       "      <td>2020</td>\n",
       "      <td>Kolbotn</td>\n",
       "      <td>Sandviken</td>\n",
       "      <td>Home</td>\n",
       "      <td>19.0</td>\n",
       "      <td>0.428571</td>\n",
       "      <td>2020-08-16</td>\n",
       "      <td>1.59</td>\n",
       "      <td>1.95</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>175</th>\n",
       "      <td>2020</td>\n",
       "      <td>Avaldsnes</td>\n",
       "      <td>Sandviken</td>\n",
       "      <td>Away</td>\n",
       "      <td>17.0</td>\n",
       "      <td>0.714286</td>\n",
       "      <td>2020-12-06</td>\n",
       "      <td>1.82</td>\n",
       "      <td>1.95</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>176</th>\n",
       "      <td>2020</td>\n",
       "      <td>Arna-Bjørnar</td>\n",
       "      <td>LSK Kvinner</td>\n",
       "      <td>Home</td>\n",
       "      <td>15.0</td>\n",
       "      <td>-1.285714</td>\n",
       "      <td>2020-10-03</td>\n",
       "      <td>1.18</td>\n",
       "      <td>2.27</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>177</th>\n",
       "      <td>2020</td>\n",
       "      <td>Avaldsnes</td>\n",
       "      <td>Lyn</td>\n",
       "      <td>Home</td>\n",
       "      <td>18.0</td>\n",
       "      <td>1.200000</td>\n",
       "      <td>2020-07-04</td>\n",
       "      <td>1.82</td>\n",
       "      <td>1.05</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>178</th>\n",
       "      <td>2020</td>\n",
       "      <td>Arna-Bjørnar</td>\n",
       "      <td>Sandviken</td>\n",
       "      <td>Home</td>\n",
       "      <td>16.0</td>\n",
       "      <td>-0.666667</td>\n",
       "      <td>2020-08-26</td>\n",
       "      <td>1.18</td>\n",
       "      <td>1.95</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>179</th>\n",
       "      <td>2020</td>\n",
       "      <td>Arna-Bjørnar</td>\n",
       "      <td>Sandviken</td>\n",
       "      <td>Away</td>\n",
       "      <td>17.0</td>\n",
       "      <td>-0.571429</td>\n",
       "      <td>2020-09-27</td>\n",
       "      <td>1.18</td>\n",
       "      <td>1.95</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>180</th>\n",
       "      <td>2020</td>\n",
       "      <td>Røa</td>\n",
       "      <td>LSK Kvinner</td>\n",
       "      <td>Home</td>\n",
       "      <td>18.0</td>\n",
       "      <td>-2.000000</td>\n",
       "      <td>2020-07-17</td>\n",
       "      <td>1.77</td>\n",
       "      <td>2.27</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>181</th>\n",
       "      <td>2020</td>\n",
       "      <td>Kolbotn</td>\n",
       "      <td>LSK Kvinner</td>\n",
       "      <td>Home</td>\n",
       "      <td>20.0</td>\n",
       "      <td>-1.714286</td>\n",
       "      <td>2020-11-07</td>\n",
       "      <td>1.59</td>\n",
       "      <td>2.27</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>182</th>\n",
       "      <td>2020</td>\n",
       "      <td>Arna-Bjørnar</td>\n",
       "      <td>Røa</td>\n",
       "      <td>Away</td>\n",
       "      <td>18.0</td>\n",
       "      <td>-0.571429</td>\n",
       "      <td>2020-07-04</td>\n",
       "      <td>1.18</td>\n",
       "      <td>1.77</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>183</th>\n",
       "      <td>2020</td>\n",
       "      <td>Sandviken</td>\n",
       "      <td>Lyn</td>\n",
       "      <td>Away</td>\n",
       "      <td>18.0</td>\n",
       "      <td>1.800000</td>\n",
       "      <td>2020-09-12</td>\n",
       "      <td>1.95</td>\n",
       "      <td>1.05</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     Season          Team     Opponent Venue  Game # in season  \\\n",
       "174    2020       Kolbotn    Sandviken  Home              19.0   \n",
       "175    2020     Avaldsnes    Sandviken  Away              17.0   \n",
       "176    2020  Arna-Bjørnar  LSK Kvinner  Home              15.0   \n",
       "177    2020     Avaldsnes          Lyn  Home              18.0   \n",
       "178    2020  Arna-Bjørnar    Sandviken  Home              16.0   \n",
       "179    2020  Arna-Bjørnar    Sandviken  Away              17.0   \n",
       "180    2020           Røa  LSK Kvinner  Home              18.0   \n",
       "181    2020       Kolbotn  LSK Kvinner  Home              20.0   \n",
       "182    2020  Arna-Bjørnar          Røa  Away              18.0   \n",
       "183    2020     Sandviken          Lyn  Away              18.0   \n",
       "\n",
       "     Avg. Historical Score Diff       Date  \\\n",
       "174                    0.428571 2020-08-16   \n",
       "175                    0.714286 2020-12-06   \n",
       "176                   -1.285714 2020-10-03   \n",
       "177                    1.200000 2020-07-04   \n",
       "178                   -0.666667 2020-08-26   \n",
       "179                   -0.571429 2020-09-27   \n",
       "180                   -2.000000 2020-07-17   \n",
       "181                   -1.714286 2020-11-07   \n",
       "182                   -0.571429 2020-07-04   \n",
       "183                    1.800000 2020-09-12   \n",
       "\n",
       "     Team's goals per 90 min last season  \\\n",
       "174                                 1.59   \n",
       "175                                 1.82   \n",
       "176                                 1.18   \n",
       "177                                 1.82   \n",
       "178                                 1.18   \n",
       "179                                 1.18   \n",
       "180                                 1.77   \n",
       "181                                 1.59   \n",
       "182                                 1.18   \n",
       "183                                 1.95   \n",
       "\n",
       "     Opponent's goals per 90 min last season  y1: Team Score  \\\n",
       "174                                     1.95             NaN   \n",
       "175                                     1.95             NaN   \n",
       "176                                     2.27             NaN   \n",
       "177                                     1.05             NaN   \n",
       "178                                     1.95             NaN   \n",
       "179                                     1.95             NaN   \n",
       "180                                     2.27             NaN   \n",
       "181                                     2.27             NaN   \n",
       "182                                     1.77             NaN   \n",
       "183                                     1.05             NaN   \n",
       "\n",
       "     y2: Opponent Score  \n",
       "174                 NaN  \n",
       "175                 NaN  \n",
       "176                 NaN  \n",
       "177                 NaN  \n",
       "178                 NaN  \n",
       "179                 NaN  \n",
       "180                 NaN  \n",
       "181                 NaN  \n",
       "182                 NaN  \n",
       "183                 NaN  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Creating two new features with (1) the avg. number of goals the team and (2) the opponent scored last season per 90 min.\n",
    "\n",
    "# Calculating the mean number of scored goals for a team in the 2017 season (for imputation purposes)\n",
    "goals_in_2017 = games_df.loc[(games_df[\"Season\"] == 2017)][\"y1: Team Score\"]\n",
    "avg_goals_in_2017 = np.mean(goals_in_2017)\n",
    "print(avg_goals_in_2017)\n",
    "\n",
    "# Calculating one std below the mean of scored goals for a team in each season (for imputation purposes)\n",
    "goals_2017_low_avg_score = avg_goals_in_2017 - np.std(goals_in_2017) * 0.5\n",
    "print(f\"Half a standard deviation from the mean of goals scored for a single team in the 2017 season is {goals_2017_low_avg_score}\")\n",
    "\n",
    "goals_in_2018 = games_df.loc[(games_df[\"Season\"] == 2018)][\"y1: Team Score\"]\n",
    "goals_2018_low_avg_score = np.mean(goals_in_2018) - np.std(goals_in_2018) * 0.5\n",
    "print(f\"Half a standard deviation from the mean of goals scored for single team in the 2018 season is {goals_2018_low_avg_score}\")\n",
    "\n",
    "goals_in_2019 = games_df.loc[(games_df[\"Season\"] == 2019)][\"y1: Team Score\"]\n",
    "goals_2019_low_avg_score = np.mean(goals_in_2019) - np.std(goals_in_2019) * 0.5\n",
    "print(f\"Half a standard deviation from the mean of goals scored for a single team in the 2019 season is {goals_2019_low_avg_score}\")\n",
    "\n",
    "\n",
    "team_goals_per_90_min_last_season = []\n",
    "opponent_goals_per_90_min_last_season = []\n",
    "\n",
    "# Looping over games_df and retrieving values from teams_df and adding them to the calculated results as I go\n",
    "for index, row in games_df.iterrows():\n",
    "    team = row[\"Team\"]\n",
    "    opponent = row[\"Opponent\"]\n",
    "    current_season = row[\"Season\"]      \n",
    "    \n",
    "    # There are no data to use for the first season, so I'll impute the average score (for a single team) for matches in the 2017 season \n",
    "    if current_season == 2017:\n",
    "        team_goals_per_90_min_last_season.append(avg_goals_in_2017)\n",
    "        opponent_goals_per_90_min_last_season.append(avg_goals_in_2017)\n",
    "    \n",
    "    # For the 2018 season I get the avg. goals per 90 min from the data for the 2017 season\n",
    "    # A few teams didn't play in Toppserien the previous year, so some exception handling is needed\n",
    "    elif current_season == 2018:\n",
    "        try:\n",
    "            team_avg_gls_season_2017 = teams_2017_df.loc[(teams_2017_df[('Unnamed: 0_level_0',  'Squad')] == team), \n",
    "                                                         ('Per 90 Minutes', 'Gls')].iloc[0]\n",
    "            team_goals_per_90_min_last_season.append(team_avg_gls_season_2017)\n",
    "        except IndexError:\n",
    "            team_goals_per_90_min_last_season.append(goals_2017_low_avg_score)  # imputing a low avg score\n",
    "            \n",
    "        try:\n",
    "            opponent_avg_gls_season_2017 = teams_2017_df.loc[(teams_2017_df[('Unnamed: 0_level_0',  'Squad')] == opponent), \n",
    "                                                             ('Per 90 Minutes', 'Gls')].iloc[0]\n",
    "            opponent_goals_per_90_min_last_season.append(opponent_avg_gls_season_2017)\n",
    "        except IndexError:\n",
    "            opponent_goals_per_90_min_last_season.append(goals_2017_low_avg_score)  # impute a low avg score\n",
    "\n",
    "    elif current_season == 2019:\n",
    "        try:\n",
    "            team_avg_gls_season_2018 = teams_2018_df.loc[(teams_2018_df[('Unnamed: 0_level_0',  'Squad')] == team), \n",
    "                                                         ('Per 90 Minutes', 'Gls')].iloc[0]\n",
    "            team_goals_per_90_min_last_season.append(team_avg_gls_season_2018)\n",
    "        except IndexError:\n",
    "            team_goals_per_90_min_last_season.append(goals_2018_low_avg_score)  # impute a low avg score\n",
    "            \n",
    "        try:\n",
    "            opponent_avg_gls_season_2018 = teams_2018_df.loc[(teams_2018_df[('Unnamed: 0_level_0',  'Squad')] == opponent), \n",
    "                                                             ('Per 90 Minutes', 'Gls')].iloc[0]\n",
    "            opponent_goals_per_90_min_last_season.append(opponent_avg_gls_season_2018)\n",
    "        except IndexError:\n",
    "            opponent_goals_per_90_min_last_season.append(goals_2018_low_avg_score)  # impute a low avg score\n",
    "\n",
    "    elif current_season == 2020:\n",
    "        try:\n",
    "            team_avg_gls_season_2019 = teams_2019_df.loc[(teams_2019_df[('Unnamed: 0_level_0',  'Squad')] == team), \n",
    "                                                     ('Per 90 Minutes', 'Gls')].iloc[0]\n",
    "            team_goals_per_90_min_last_season.append(team_avg_gls_season_2019)\n",
    "        except IndexError:\n",
    "            team_goals_per_90_min_last_season.append(goals_2019_low_avg_score)  # impute a low avg score\n",
    "            \n",
    "        try:\n",
    "            opponent_avg_gls_season_2019 = teams_2019_df.loc[(teams_2019_df[('Unnamed: 0_level_0',  'Squad')] == opponent), \n",
    "                                                             ('Per 90 Minutes', 'Gls')].iloc[0]\n",
    "            opponent_goals_per_90_min_last_season.append(opponent_avg_gls_season_2019)\n",
    "        except IndexError:\n",
    "            opponent_goals_per_90_min_last_season.append(goals_2019_low_avg_score)  # impute a low avg score\n",
    "\n",
    "# Adding the two new features\n",
    "games_df[\"Team's goals per 90 min last season\"] = team_goals_per_90_min_last_season\n",
    "games_df[\"Opponent's goals per 90 min last season\"] = opponent_goals_per_90_min_last_season\n",
    "\n",
    "# Reordering the columns again so that the dependent variables still appear at the end\n",
    "# Reordering the features so that the predictor variables appear at the end\n",
    "games_df = games_df[[col for col in games_df.columns \n",
    "                     if col not in [\"y1: Team Score\", \"y2: Opponent Score\"]] + [\"y1: Team Score\", \"y2: Opponent Score\"]]\n",
    "\n",
    "# displaying the resulting dataframe\n",
    "display(games_df.head(10))\n",
    "display(games_df.tail(10))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0188e21e",
   "metadata": {},
   "source": [
    "[*back to top*](#top)\n",
    "### <a id=\"team-avg-age\"></a>3.8) Adding the teams' average age"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1a505e36",
   "metadata": {},
   "source": [
    "Now I am going to add the average age of the team's players, as well as the average age of the opponent team's players. I honestly don't think this is going to make much of a difference, but I will add it in here anyway and then check the correlation with the dependent variables later (though I still doubt there is causation). It is however interesting to note that the difference in average age for the \"oldest\" and \"youngest\" teams is 6.6 years, which is a lot more than I expected. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "123595d9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Player</th>\n",
       "      <th>Nation</th>\n",
       "      <th>Pos</th>\n",
       "      <th>Squad</th>\n",
       "      <th>Age</th>\n",
       "      <th>Born</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Kamilla Aabel</td>\n",
       "      <td>no NOR</td>\n",
       "      <td>DF</td>\n",
       "      <td>Røa</td>\n",
       "      <td>23.0</td>\n",
       "      <td>1996.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Anna Aahjem</td>\n",
       "      <td>no NOR</td>\n",
       "      <td>MF</td>\n",
       "      <td>Lyn</td>\n",
       "      <td>20.0</td>\n",
       "      <td>2000.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Meryll Abrahamsen</td>\n",
       "      <td>no NOR</td>\n",
       "      <td>DF</td>\n",
       "      <td>Arna-Bjørnar</td>\n",
       "      <td>22.0</td>\n",
       "      <td>1997.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Rasheedat Ajibade</td>\n",
       "      <td>ng NGA</td>\n",
       "      <td>FW</td>\n",
       "      <td>Avaldsnes</td>\n",
       "      <td>20.0</td>\n",
       "      <td>1999.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Marija Aleksić</td>\n",
       "      <td>ba BIH</td>\n",
       "      <td>MF</td>\n",
       "      <td>Arna-Bjørnar</td>\n",
       "      <td>22.0</td>\n",
       "      <td>1997.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              Player  Nation Pos         Squad   Age    Born\n",
       "1      Kamilla Aabel  no NOR  DF           Røa  23.0  1996.0\n",
       "2        Anna Aahjem  no NOR  MF           Lyn  20.0  2000.0\n",
       "3  Meryll Abrahamsen  no NOR  DF  Arna-Bjørnar  22.0  1997.0\n",
       "4  Rasheedat Ajibade  ng NGA  FW     Avaldsnes  20.0  1999.0\n",
       "5     Marija Aleksić  ba BIH  MF  Arna-Bjørnar  22.0  1997.0"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Loading the player_stats dataset for the 2020 season\n",
    "players_2020_df = pd.read_html(\"Raw Data/2020/player-stats.xls\", encoding=\"UTF-8\")[0]\n",
    "players_2020_df.head()\n",
    "\n",
    "# Drop rows where all values are missing\n",
    "players_2020_df = players_2020_df.dropna(how='all')\n",
    "players_2020_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "55bd0057",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The average of average team ages for 2017 is: 23.44\n",
      "The average of average team ages for 2018 is: 23.3\n",
      "The average of average team ages for 2019 is: 23.41\n"
     ]
    }
   ],
   "source": [
    "# Calculating the average of average ages for the teams in each season (for imputation purposes)\n",
    "# NB! Not actually the same as the average age, because teams may have a different nr of players\n",
    "avg_ages_2017 = teams_2017_df[('Unnamed: 2_level_0', 'Age')].values\n",
    "avg_avg_ages_2017 = avg_ages_2017.sum() / len(avg_ages_2017)\n",
    "print(f\"The average of average team ages for 2017 is: {round(avg_avg_ages_2017, 2)}\")\n",
    "\n",
    "avg_ages_2018 = teams_2018_df[('Unnamed: 2_level_0', 'Age')].values\n",
    "avg_avg_ages_2018 = avg_ages_2018.sum() / len(avg_ages_2017)\n",
    "print(f\"The average of average team ages for 2018 is: {round(avg_avg_ages_2018, 2)}\")\n",
    "\n",
    "avg_ages_2019 = teams_2019_df[('Unnamed: 2_level_0', 'Age')].values\n",
    "avg_avg_ages_2019 = avg_ages_2019.sum() / len(avg_ages_2017)\n",
    "print(f\"The average of average team ages for 2019 is: {round(avg_avg_ages_2019, 2)}\")\n",
    "\n",
    "# Calculating the average of average ages for the teams in the 2020 season (for imputation purposes)\n",
    "team_player_ages_2020 = defaultdict(list)\n",
    "\n",
    "# For each player in the 2020 season\n",
    "for index, row in players_2020_df.iterrows():\n",
    "    age = row[\"Age\"]\n",
    "    team = row[\"Squad\"]\n",
    "    # Handling entries with missing values by not using them for calculations\n",
    "    if not pd.isnull(team) and not pd.isnull(age):\n",
    "        # Add the player's age to the dict of the player's team's ages\n",
    "        team_player_ages_2020[team].append(age)\n",
    "\n",
    "# Performing the calculations\n",
    "avg_team_ages_2020 = [sum(values) / len(values) for values in team_player_ages_2020.values()]\n",
    "avg_avg_ages_2020 = sum(avg_team_ages_2020) / len(avg_team_ages_2020)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "6ff44679",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Season</th>\n",
       "      <th>Team</th>\n",
       "      <th>Opponent</th>\n",
       "      <th>Venue</th>\n",
       "      <th>Game # in season</th>\n",
       "      <th>Avg. Historical Score Diff</th>\n",
       "      <th>Date</th>\n",
       "      <th>Team's goals per 90 min last season</th>\n",
       "      <th>Opponent's goals per 90 min last season</th>\n",
       "      <th>Team's Avg. Age</th>\n",
       "      <th>Opponent's Avg. Age</th>\n",
       "      <th>y1: Team Score</th>\n",
       "      <th>y2: Opponent Score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2017</td>\n",
       "      <td>Medkila</td>\n",
       "      <td>Sandviken</td>\n",
       "      <td>Home</td>\n",
       "      <td>1.0</td>\n",
       "      <td>-0.166667</td>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>1.640152</td>\n",
       "      <td>1.640152</td>\n",
       "      <td>21.9</td>\n",
       "      <td>25.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2017</td>\n",
       "      <td>Arna-Bjørnar</td>\n",
       "      <td>Grand Bodø</td>\n",
       "      <td>Away</td>\n",
       "      <td>1.0</td>\n",
       "      <td>-0.166667</td>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>1.640152</td>\n",
       "      <td>1.640152</td>\n",
       "      <td>22.9</td>\n",
       "      <td>25.2</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2017</td>\n",
       "      <td>Stabæk</td>\n",
       "      <td>Røa</td>\n",
       "      <td>Away</td>\n",
       "      <td>2.0</td>\n",
       "      <td>-1.833333</td>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>1.640152</td>\n",
       "      <td>1.640152</td>\n",
       "      <td>21.9</td>\n",
       "      <td>22.4</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2017</td>\n",
       "      <td>Klepp</td>\n",
       "      <td>LSK Kvinner</td>\n",
       "      <td>Away</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.250000</td>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>1.640152</td>\n",
       "      <td>1.640152</td>\n",
       "      <td>25.2</td>\n",
       "      <td>20.1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>6.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2017</td>\n",
       "      <td>Kolbotn</td>\n",
       "      <td>Trondheims-Ørn</td>\n",
       "      <td>Away</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.166667</td>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>1.640152</td>\n",
       "      <td>1.640152</td>\n",
       "      <td>22.4</td>\n",
       "      <td>21.8</td>\n",
       "      <td>2.0</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Season          Team        Opponent Venue  Game # in season  \\\n",
       "0    2017       Medkila       Sandviken  Home               1.0   \n",
       "1    2017  Arna-Bjørnar      Grand Bodø  Away               1.0   \n",
       "2    2017        Stabæk             Røa  Away               2.0   \n",
       "3    2017         Klepp     LSK Kvinner  Away               1.0   \n",
       "4    2017       Kolbotn  Trondheims-Ørn  Away               1.0   \n",
       "\n",
       "   Avg. Historical Score Diff       Date  Team's goals per 90 min last season  \\\n",
       "0                   -0.166667 2017-04-17                             1.640152   \n",
       "1                   -0.166667 2017-04-17                             1.640152   \n",
       "2                   -1.833333 2017-04-17                             1.640152   \n",
       "3                    1.250000 2017-04-17                             1.640152   \n",
       "4                    1.166667 2017-04-17                             1.640152   \n",
       "\n",
       "   Opponent's goals per 90 min last season  Team's Avg. Age  \\\n",
       "0                                 1.640152             21.9   \n",
       "1                                 1.640152             22.9   \n",
       "2                                 1.640152             21.9   \n",
       "3                                 1.640152             25.2   \n",
       "4                                 1.640152             22.4   \n",
       "\n",
       "   Opponent's Avg. Age  y1: Team Score  y2: Opponent Score  \n",
       "0                 25.0             1.0                 1.0  \n",
       "1                 25.2             2.0                 2.0  \n",
       "2                 22.4             2.0                 1.0  \n",
       "3                 20.1             1.0                 6.0  \n",
       "4                 21.8             2.0                 3.0  "
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Create two new features for team's and opponent's average age\n",
    "for index, row in games_df.iterrows():\n",
    "    team = row[\"Team\"]\n",
    "    opponent = row[\"Opponent\"]\n",
    "    current_season = row[\"Season\"]\n",
    "    \n",
    "    if current_season == 2017:\n",
    "        team_avg_age = teams_2017_df.loc[(teams_2017_df[('Unnamed: 0_level_0',  'Squad')] == team), \n",
    "                                         ('Unnamed: 2_level_0', 'Age')]\n",
    "        opponent_avg_age = teams_2017_df.loc[(teams_2017_df[('Unnamed: 0_level_0',  'Squad')] == opponent), \n",
    "                                             ('Unnamed: 2_level_0', 'Age')]\n",
    "        \n",
    "        try:\n",
    "            games_df.loc[[index], [\"Team's Avg. Age\"]] = float(team_avg_age)  \n",
    "        except TypeError as err:\n",
    "            games_df.loc[[index], [\"Team's Avg. Age\"]] = avg_avg_ages_2017  # Impute avg of avg ages 2017\n",
    "            \n",
    "        try:\n",
    "            games_df.loc[[index], [\"Opponent's Avg. Age\"]] = float(opponent_avg_age)\n",
    "        except TypeError as err:\n",
    "            games_df.loc[[index], [\"Opponent's Avg. Age\"]] = avg_avg_ages_2017  # Impute avg of avg ages 2017\n",
    "            \n",
    "                     \n",
    "    elif current_season == 2018:\n",
    "        team_avg_age = teams_2018_df.loc[(teams_2018_df[('Unnamed: 0_level_0',  'Squad')] == team), \n",
    "                                         ('Unnamed: 2_level_0', 'Age')]\n",
    "        opponent_avg_age = teams_2018_df.loc[(teams_2018_df[('Unnamed: 0_level_0',  'Squad')] == opponent), \n",
    "                                             ('Unnamed: 2_level_0', 'Age')]\n",
    "        \n",
    "        try:\n",
    "            games_df.loc[[index], [\"Team's Avg. Age\"]] = float(team_avg_age)  \n",
    "        except TypeError as err:\n",
    "            games_df.loc[[index], [\"Team's Avg. Age\"]] = avg_avg_ages_2018  # Impute avg of avg ages 2018\n",
    "            \n",
    "        try:\n",
    "            games_df.loc[[index], [\"Opponent's Avg. Age\"]] = float(opponent_avg_age)\n",
    "        except TypeError as err:\n",
    "            games_df.loc[[index], [\"Opponent's Avg. Age\"]] = avg_avg_ages_2018  # Impute avg of avg ages 2018\n",
    "\n",
    "            \n",
    "    elif current_season == 2019:\n",
    "        team_avg_age = teams_2019_df.loc[(teams_2019_df[('Unnamed: 0_level_0',  'Squad')] == team), \n",
    "                                         ('Unnamed: 2_level_0', 'Age')]\n",
    "        opponent_avg_age = teams_2019_df.loc[(teams_2019_df[('Unnamed: 0_level_0',  'Squad')] == opponent), \n",
    "                                             ('Unnamed: 2_level_0', 'Age')]\n",
    "        \n",
    "        try:\n",
    "            games_df.loc[[index], [\"Team's Avg. Age\"]] = float(team_avg_age)  \n",
    "        except TypeError as err:\n",
    "            games_df.loc[[index], [\"Team's Avg. Age\"]] = avg_avg_ages_2019  # Impute avg of avg ages 2019\n",
    "            \n",
    "        try:\n",
    "            games_df.loc[[index], [\"Opponent's Avg. Age\"]] = float(opponent_avg_age)\n",
    "        except TypeError as err:\n",
    "            games_df.loc[[index], [\"Opponent's Avg. Age\"]] = avg_avg_ages_2019  # Impute avg of avg ages 2019\n",
    "            \n",
    "    elif current_season == 2020:\n",
    "        team_avg_age = teams_2019_df.loc[(teams_2019_df[('Unnamed: 0_level_0',  'Squad')] == team), \n",
    "                                         ('Unnamed: 2_level_0', 'Age')]\n",
    "        opponent_avg_age = teams_2019_df.loc[(teams_2019_df[('Unnamed: 0_level_0',  'Squad')] == opponent), \n",
    "                                             ('Unnamed: 2_level_0', 'Age')]\n",
    "        \n",
    "        try:\n",
    "            games_df.loc[[index], [\"Team's Avg. Age\"]] = float(team_avg_age)  \n",
    "        except TypeError as err:\n",
    "            games_df.loc[[index], [\"Team's Avg. Age\"]] = avg_avg_ages_2020  # Impute avg of avg ages 2019\n",
    "            \n",
    "        try:\n",
    "            games_df.loc[[index], [\"Opponent's Avg. Age\"]] = float(opponent_avg_age)\n",
    "        except TypeError as err:\n",
    "            games_df.loc[[index], [\"Opponent's Avg. Age\"]] = avg_avg_ages_2020  # Impute avg of avg ages 2019\n",
    "  \n",
    "            \n",
    "# Reordering the features so that the predictor variables appear at the end\n",
    "games_df = games_df[[col for col in games_df.columns if col not in [\"y1: Team Score\", \"y2: Opponent Score\"]] + [\"y1: Team Score\", \"y2: Opponent Score\"]]\n",
    "\n",
    "games_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "09cf71ea",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The range of avg. ages is from 19.8 years to 26.4 years, a difference of 6.6 years.\n"
     ]
    }
   ],
   "source": [
    "# Calculate the avg. age difference between \"oldest\" and \"youngest\" teams\n",
    "avg_ages = set(games_df[\"Team's Avg. Age\"].values)\n",
    "print(f\"The range of avg. ages is from {min(avg_ages)} years to {max(avg_ages)} years, a difference of {round(max(avg_ages) - min(avg_ages), 2)} years.\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "d02e13e1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 936x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Visualizing the distribution of avg. ages\n",
    "teams_avg_age_2017 = teams_2017_df[('Unnamed: 2_level_0', 'Age')]\n",
    "teams_avg_age_2018 = teams_2017_df[('Unnamed: 2_level_0', 'Age')]\n",
    "teams_avg_age_2019 = teams_2017_df[('Unnamed: 2_level_0', 'Age')]\n",
    "teams_avg_ages_all_seasons = teams_avg_age_2017.append([teams_avg_age_2018, teams_avg_age_2019]).values\n",
    "    \n",
    "fig, ax1 = plt.subplots()\n",
    "n, bins, patches = ax1.hist(teams_avg_ages_all_seasons, label='Team Avg. Ages across all seasons',\n",
    "                            stacked=False, color='darkred', bins=12)\n",
    "# Set labels for histogram\n",
    "ax1.set_ylabel('Count', size=12)\n",
    "ax1.set_xlabel(\"Team's avg. age\", size=12)\n",
    "ax1.set_title('Distibution of avg. team age for the Toppserien soccer championships, 2017 - 2019', size=18)\n",
    "ax1.legend(bbox_to_anchor=(1, 0.92))\n",
    "\n",
    "fig.set_figheight(8)\n",
    "fig.set_figwidth(13)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3c8f951c",
   "metadata": {},
   "source": [
    "### <a id=\"top-tier-team\"></a>3.9) Adding a feature to distinguish top-tier teams"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d532cc62",
   "metadata": {},
   "source": [
    "Finally, I want to add a dummy variable that indicates whether the teams finished in the top 5 last season or not. \n",
    "This should work as an additional indicator that hopefully helps the model understand that some teams are \"star teams\" and perform a lot better than the rest. We can for example see from the distribution of scores that *LSK Kvinner* and *Vålerenga* have been dominating Toppserien for the last three years. Later I will experiment more with this metric, in order to find the optimum value for n for the \"top n\" teams that dominate the championships. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "solar-glass",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Rk</th>\n",
       "      <th>Squad</th>\n",
       "      <th>MP</th>\n",
       "      <th>W</th>\n",
       "      <th>D</th>\n",
       "      <th>L</th>\n",
       "      <th>GF</th>\n",
       "      <th>GA</th>\n",
       "      <th>GD</th>\n",
       "      <th>Pts</th>\n",
       "      <th>Attendance</th>\n",
       "      <th>Top Team Scorer</th>\n",
       "      <th>Goalkeeper</th>\n",
       "      <th>Notes</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>LSK Kvinner</td>\n",
       "      <td>22</td>\n",
       "      <td>15</td>\n",
       "      <td>5</td>\n",
       "      <td>2</td>\n",
       "      <td>51</td>\n",
       "      <td>18</td>\n",
       "      <td>33</td>\n",
       "      <td>50</td>\n",
       "      <td>339</td>\n",
       "      <td>Guro Reiten, Elise Thorsnes - 12</td>\n",
       "      <td>Cecilie Fiskerstrand</td>\n",
       "      <td>→ Champions League via league finish</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>Vålerenga</td>\n",
       "      <td>22</td>\n",
       "      <td>14</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>41</td>\n",
       "      <td>24</td>\n",
       "      <td>17</td>\n",
       "      <td>46</td>\n",
       "      <td>489</td>\n",
       "      <td>Natasha Dowie, Ajara Nchout - 11</td>\n",
       "      <td>Hannah Seabert</td>\n",
       "      <td>→ Champions League via league finish</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>Klepp</td>\n",
       "      <td>22</td>\n",
       "      <td>14</td>\n",
       "      <td>2</td>\n",
       "      <td>6</td>\n",
       "      <td>48</td>\n",
       "      <td>19</td>\n",
       "      <td>29</td>\n",
       "      <td>44</td>\n",
       "      <td>260</td>\n",
       "      <td>Havana Solaun, Tameka Butt - 10</td>\n",
       "      <td>Lindsey Harris</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>Sandviken</td>\n",
       "      <td>22</td>\n",
       "      <td>12</td>\n",
       "      <td>4</td>\n",
       "      <td>6</td>\n",
       "      <td>43</td>\n",
       "      <td>32</td>\n",
       "      <td>11</td>\n",
       "      <td>40</td>\n",
       "      <td>465</td>\n",
       "      <td>Kennya Cordner - 17</td>\n",
       "      <td>Nora Neset Gjøen</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>Avaldsnes</td>\n",
       "      <td>22</td>\n",
       "      <td>9</td>\n",
       "      <td>7</td>\n",
       "      <td>6</td>\n",
       "      <td>40</td>\n",
       "      <td>33</td>\n",
       "      <td>7</td>\n",
       "      <td>34</td>\n",
       "      <td>172</td>\n",
       "      <td>Giovanna - 9</td>\n",
       "      <td>Victoria Esson</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>6</td>\n",
       "      <td>Røa</td>\n",
       "      <td>22</td>\n",
       "      <td>7</td>\n",
       "      <td>8</td>\n",
       "      <td>7</td>\n",
       "      <td>40</td>\n",
       "      <td>35</td>\n",
       "      <td>5</td>\n",
       "      <td>29</td>\n",
       "      <td>305</td>\n",
       "      <td>Synne Jensen - 16</td>\n",
       "      <td>Kirvil Schau Odden</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>7</td>\n",
       "      <td>Trondheims-Ørn</td>\n",
       "      <td>22</td>\n",
       "      <td>8</td>\n",
       "      <td>5</td>\n",
       "      <td>9</td>\n",
       "      <td>26</td>\n",
       "      <td>22</td>\n",
       "      <td>4</td>\n",
       "      <td>29</td>\n",
       "      <td>264</td>\n",
       "      <td>Marit Clausen - 9</td>\n",
       "      <td>Kristine Nøstmo</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>8</td>\n",
       "      <td>Kolbotn</td>\n",
       "      <td>22</td>\n",
       "      <td>7</td>\n",
       "      <td>6</td>\n",
       "      <td>9</td>\n",
       "      <td>35</td>\n",
       "      <td>39</td>\n",
       "      <td>-4</td>\n",
       "      <td>27</td>\n",
       "      <td>167</td>\n",
       "      <td>Isabell Herlovsen - 8</td>\n",
       "      <td>Aurora Mikalsen</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>9</td>\n",
       "      <td>Arna-Bjørnar</td>\n",
       "      <td>22</td>\n",
       "      <td>6</td>\n",
       "      <td>5</td>\n",
       "      <td>11</td>\n",
       "      <td>26</td>\n",
       "      <td>41</td>\n",
       "      <td>-15</td>\n",
       "      <td>23</td>\n",
       "      <td>295</td>\n",
       "      <td>Milica Mijatović - 7</td>\n",
       "      <td>Oda Bogstad</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>10</td>\n",
       "      <td>Lyn</td>\n",
       "      <td>22</td>\n",
       "      <td>4</td>\n",
       "      <td>8</td>\n",
       "      <td>10</td>\n",
       "      <td>23</td>\n",
       "      <td>37</td>\n",
       "      <td>-14</td>\n",
       "      <td>20</td>\n",
       "      <td>191</td>\n",
       "      <td>Runa Lillegård - 10</td>\n",
       "      <td>Karen Oline Sneve</td>\n",
       "      <td>→ Norwegian 1/2 Relegation/Promotion Playoff (W)</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>11</td>\n",
       "      <td>Stabæk</td>\n",
       "      <td>22</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>15</td>\n",
       "      <td>16</td>\n",
       "      <td>43</td>\n",
       "      <td>-27</td>\n",
       "      <td>13</td>\n",
       "      <td>180</td>\n",
       "      <td>Melissa Bjånesøy - 9</td>\n",
       "      <td>Ingrid Hjelmseth</td>\n",
       "      <td>Relegated</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>12</td>\n",
       "      <td>Fart</td>\n",
       "      <td>22</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "      <td>16</td>\n",
       "      <td>14</td>\n",
       "      <td>60</td>\n",
       "      <td>-46</td>\n",
       "      <td>10</td>\n",
       "      <td>236</td>\n",
       "      <td>Julie Blakstad - 4</td>\n",
       "      <td>Lena Lauve</td>\n",
       "      <td>Relegated</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    Rk           Squad  MP   W  D   L  GF  GA  GD  Pts  Attendance  \\\n",
       "0    1     LSK Kvinner  22  15  5   2  51  18  33   50         339   \n",
       "1    2       Vålerenga  22  14  4   4  41  24  17   46         489   \n",
       "2    3           Klepp  22  14  2   6  48  19  29   44         260   \n",
       "3    4       Sandviken  22  12  4   6  43  32  11   40         465   \n",
       "4    5       Avaldsnes  22   9  7   6  40  33   7   34         172   \n",
       "5    6             Røa  22   7  8   7  40  35   5   29         305   \n",
       "6    7  Trondheims-Ørn  22   8  5   9  26  22   4   29         264   \n",
       "7    8         Kolbotn  22   7  6   9  35  39  -4   27         167   \n",
       "8    9    Arna-Bjørnar  22   6  5  11  26  41 -15   23         295   \n",
       "9   10             Lyn  22   4  8  10  23  37 -14   20         191   \n",
       "10  11          Stabæk  22   3  4  15  16  43 -27   13         180   \n",
       "11  12            Fart  22   2  4  16  14  60 -46   10         236   \n",
       "\n",
       "                     Top Team Scorer            Goalkeeper  \\\n",
       "0   Guro Reiten, Elise Thorsnes - 12  Cecilie Fiskerstrand   \n",
       "1   Natasha Dowie, Ajara Nchout - 11        Hannah Seabert   \n",
       "2    Havana Solaun, Tameka Butt - 10        Lindsey Harris   \n",
       "3                Kennya Cordner - 17      Nora Neset Gjøen   \n",
       "4                       Giovanna - 9        Victoria Esson   \n",
       "5                  Synne Jensen - 16    Kirvil Schau Odden   \n",
       "6                  Marit Clausen - 9       Kristine Nøstmo   \n",
       "7              Isabell Herlovsen - 8       Aurora Mikalsen   \n",
       "8               Milica Mijatović - 7           Oda Bogstad   \n",
       "9                Runa Lillegård - 10     Karen Oline Sneve   \n",
       "10              Melissa Bjånesøy - 9      Ingrid Hjelmseth   \n",
       "11                Julie Blakstad - 4            Lena Lauve   \n",
       "\n",
       "                                               Notes  \n",
       "0               → Champions League via league finish  \n",
       "1               → Champions League via league finish  \n",
       "2                                                NaN  \n",
       "3                                                NaN  \n",
       "4                                                NaN  \n",
       "5                                                NaN  \n",
       "6                                                NaN  \n",
       "7                                                NaN  \n",
       "8                                                NaN  \n",
       "9   → Norwegian 1/2 Relegation/Promotion Playoff (W)  \n",
       "10                                         Relegated  \n",
       "11                                         Relegated  "
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Loading the championship dataset\n",
    "table_2017_df = pd.read_html(\"Raw Data/2017/table.xls\", encoding=\"UTF-8\")[0]\n",
    "table_2018_df = pd.read_html(\"Raw Data/2018/table.xls\", encoding=\"UTF-8\")[0]\n",
    "table_2019_df = pd.read_html(\"Raw Data/2019/table.xls\", encoding=\"UTF-8\")[0]\n",
    "table_2019_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "614364f7",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Removing features irrelevant for exploration and prediction\n",
    "for df in [table_2017_df, table_2018_df, table_2019_df]:\n",
    "    df.drop(labels=\"Notes\", inplace=True, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "8e8b0229",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Season</th>\n",
       "      <th>Team</th>\n",
       "      <th>Opponent</th>\n",
       "      <th>Venue</th>\n",
       "      <th>Game # in season</th>\n",
       "      <th>Avg. Historical Score Diff</th>\n",
       "      <th>Date</th>\n",
       "      <th>Team's goals per 90 min last season</th>\n",
       "      <th>Opponent's goals per 90 min last season</th>\n",
       "      <th>Team's Avg. Age</th>\n",
       "      <th>Opponent's Avg. Age</th>\n",
       "      <th>Top 5 Team</th>\n",
       "      <th>Top 5 Opponent</th>\n",
       "      <th>y1: Team Score</th>\n",
       "      <th>y2: Opponent Score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2017</td>\n",
       "      <td>Medkila</td>\n",
       "      <td>Sandviken</td>\n",
       "      <td>Home</td>\n",
       "      <td>1.0</td>\n",
       "      <td>-0.166667</td>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>1.640152</td>\n",
       "      <td>1.640152</td>\n",
       "      <td>21.9</td>\n",
       "      <td>25.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2017</td>\n",
       "      <td>Arna-Bjørnar</td>\n",
       "      <td>Grand Bodø</td>\n",
       "      <td>Away</td>\n",
       "      <td>1.0</td>\n",
       "      <td>-0.166667</td>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>1.640152</td>\n",
       "      <td>1.640152</td>\n",
       "      <td>22.9</td>\n",
       "      <td>25.2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2017</td>\n",
       "      <td>Stabæk</td>\n",
       "      <td>Røa</td>\n",
       "      <td>Away</td>\n",
       "      <td>2.0</td>\n",
       "      <td>-1.833333</td>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>1.640152</td>\n",
       "      <td>1.640152</td>\n",
       "      <td>21.9</td>\n",
       "      <td>22.4</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2017</td>\n",
       "      <td>Klepp</td>\n",
       "      <td>LSK Kvinner</td>\n",
       "      <td>Away</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.250000</td>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>1.640152</td>\n",
       "      <td>1.640152</td>\n",
       "      <td>25.2</td>\n",
       "      <td>20.1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>6.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2017</td>\n",
       "      <td>Kolbotn</td>\n",
       "      <td>Trondheims-Ørn</td>\n",
       "      <td>Away</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.166667</td>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>1.640152</td>\n",
       "      <td>1.640152</td>\n",
       "      <td>22.4</td>\n",
       "      <td>21.8</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Season          Team        Opponent Venue  Game # in season  \\\n",
       "0    2017       Medkila       Sandviken  Home               1.0   \n",
       "1    2017  Arna-Bjørnar      Grand Bodø  Away               1.0   \n",
       "2    2017        Stabæk             Røa  Away               2.0   \n",
       "3    2017         Klepp     LSK Kvinner  Away               1.0   \n",
       "4    2017       Kolbotn  Trondheims-Ørn  Away               1.0   \n",
       "\n",
       "   Avg. Historical Score Diff       Date  Team's goals per 90 min last season  \\\n",
       "0                   -0.166667 2017-04-17                             1.640152   \n",
       "1                   -0.166667 2017-04-17                             1.640152   \n",
       "2                   -1.833333 2017-04-17                             1.640152   \n",
       "3                    1.250000 2017-04-17                             1.640152   \n",
       "4                    1.166667 2017-04-17                             1.640152   \n",
       "\n",
       "   Opponent's goals per 90 min last season  Team's Avg. Age  \\\n",
       "0                                 1.640152             21.9   \n",
       "1                                 1.640152             22.9   \n",
       "2                                 1.640152             21.9   \n",
       "3                                 1.640152             25.2   \n",
       "4                                 1.640152             22.4   \n",
       "\n",
       "   Opponent's Avg. Age  Top 5 Team  Top 5 Opponent  y1: Team Score  \\\n",
       "0                 25.0           0               0             1.0   \n",
       "1                 25.2           0               0             2.0   \n",
       "2                 22.4           0               0             2.0   \n",
       "3                 20.1           0               0             1.0   \n",
       "4                 21.8           0               0             2.0   \n",
       "\n",
       "   y2: Opponent Score  \n",
       "0                 1.0  \n",
       "1                 2.0  \n",
       "2                 1.0  \n",
       "3                 6.0  \n",
       "4                 3.0  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Season</th>\n",
       "      <th>Team</th>\n",
       "      <th>Opponent</th>\n",
       "      <th>Venue</th>\n",
       "      <th>Game # in season</th>\n",
       "      <th>Avg. Historical Score Diff</th>\n",
       "      <th>Date</th>\n",
       "      <th>Team's goals per 90 min last season</th>\n",
       "      <th>Opponent's goals per 90 min last season</th>\n",
       "      <th>Team's Avg. Age</th>\n",
       "      <th>Opponent's Avg. Age</th>\n",
       "      <th>Top 5 Team</th>\n",
       "      <th>Top 5 Opponent</th>\n",
       "      <th>y1: Team Score</th>\n",
       "      <th>y2: Opponent Score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>179</th>\n",
       "      <td>2020</td>\n",
       "      <td>Arna-Bjørnar</td>\n",
       "      <td>Sandviken</td>\n",
       "      <td>Away</td>\n",
       "      <td>17.0</td>\n",
       "      <td>-0.571429</td>\n",
       "      <td>2020-09-27</td>\n",
       "      <td>1.18</td>\n",
       "      <td>1.95</td>\n",
       "      <td>22.4</td>\n",
       "      <td>25.3</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>180</th>\n",
       "      <td>2020</td>\n",
       "      <td>Røa</td>\n",
       "      <td>LSK Kvinner</td>\n",
       "      <td>Home</td>\n",
       "      <td>18.0</td>\n",
       "      <td>-2.000000</td>\n",
       "      <td>2020-07-17</td>\n",
       "      <td>1.77</td>\n",
       "      <td>2.27</td>\n",
       "      <td>22.9</td>\n",
       "      <td>25.3</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>181</th>\n",
       "      <td>2020</td>\n",
       "      <td>Kolbotn</td>\n",
       "      <td>LSK Kvinner</td>\n",
       "      <td>Home</td>\n",
       "      <td>20.0</td>\n",
       "      <td>-1.714286</td>\n",
       "      <td>2020-11-07</td>\n",
       "      <td>1.59</td>\n",
       "      <td>2.27</td>\n",
       "      <td>21.8</td>\n",
       "      <td>25.3</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>182</th>\n",
       "      <td>2020</td>\n",
       "      <td>Arna-Bjørnar</td>\n",
       "      <td>Røa</td>\n",
       "      <td>Away</td>\n",
       "      <td>18.0</td>\n",
       "      <td>-0.571429</td>\n",
       "      <td>2020-07-04</td>\n",
       "      <td>1.18</td>\n",
       "      <td>1.77</td>\n",
       "      <td>22.4</td>\n",
       "      <td>22.9</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>183</th>\n",
       "      <td>2020</td>\n",
       "      <td>Sandviken</td>\n",
       "      <td>Lyn</td>\n",
       "      <td>Away</td>\n",
       "      <td>18.0</td>\n",
       "      <td>1.800000</td>\n",
       "      <td>2020-09-12</td>\n",
       "      <td>1.95</td>\n",
       "      <td>1.05</td>\n",
       "      <td>25.3</td>\n",
       "      <td>20.1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     Season          Team     Opponent Venue  Game # in season  \\\n",
       "179    2020  Arna-Bjørnar    Sandviken  Away              17.0   \n",
       "180    2020           Røa  LSK Kvinner  Home              18.0   \n",
       "181    2020       Kolbotn  LSK Kvinner  Home              20.0   \n",
       "182    2020  Arna-Bjørnar          Røa  Away              18.0   \n",
       "183    2020     Sandviken          Lyn  Away              18.0   \n",
       "\n",
       "     Avg. Historical Score Diff       Date  \\\n",
       "179                   -0.571429 2020-09-27   \n",
       "180                   -2.000000 2020-07-17   \n",
       "181                   -1.714286 2020-11-07   \n",
       "182                   -0.571429 2020-07-04   \n",
       "183                    1.800000 2020-09-12   \n",
       "\n",
       "     Team's goals per 90 min last season  \\\n",
       "179                                 1.18   \n",
       "180                                 1.77   \n",
       "181                                 1.59   \n",
       "182                                 1.18   \n",
       "183                                 1.95   \n",
       "\n",
       "     Opponent's goals per 90 min last season  Team's Avg. Age  \\\n",
       "179                                     1.95             22.4   \n",
       "180                                     2.27             22.9   \n",
       "181                                     2.27             21.8   \n",
       "182                                     1.77             22.4   \n",
       "183                                     1.05             25.3   \n",
       "\n",
       "     Opponent's Avg. Age  Top 5 Team  Top 5 Opponent  y1: Team Score  \\\n",
       "179                 25.3           0               1             NaN   \n",
       "180                 25.3           0               1             NaN   \n",
       "181                 25.3           0               1             NaN   \n",
       "182                 22.9           0               0             NaN   \n",
       "183                 20.1           1               0             NaN   \n",
       "\n",
       "     y2: Opponent Score  \n",
       "179                 NaN  \n",
       "180                 NaN  \n",
       "181                 NaN  \n",
       "182                 NaN  \n",
       "183                 NaN  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Adding two new features indicating whether the teams\n",
    "# finished in the top 5 last season. \n",
    "\n",
    "# .loc isn't working properly here for some reason, so I am using a list to store the values \n",
    "# for the new feature.\n",
    "top_5_team = []\n",
    "top_5_opponent = []\n",
    "\n",
    "# Looping over the games data and inserting the value for the Top5Team dummy indicators as I go\n",
    "for index, row in games_df.iterrows():\n",
    "    team = row[\"Team\"]\n",
    "    opponent = row[\"Opponent\"]\n",
    "    current_season = row[\"Season\"]\n",
    "    \n",
    "    # For 2017 values are set to 0 (no one finished in the top 5)\n",
    "    if current_season == 2017:\n",
    "        top_5_team.append(0)\n",
    "        top_5_opponent.append(0)\n",
    "        \n",
    "    \n",
    "    # For the other seasons I check what the team's placement for last year's season was\n",
    "    if current_season == 2018:\n",
    "        team_placement = table_2017_df.loc[(table_2017_df[\"Squad\"] == team), \"Rk\"]\n",
    "        opponent_placement = table_2017_df.loc[(table_2017_df[\"Squad\"] == opponent), \"Rk\"]\n",
    "        # Calculating whether the team is a top 5 team\n",
    "        try:\n",
    "            if int(team_placement) < 6:\n",
    "                top_5_team.append(1)\n",
    "            else:\n",
    "                top_5_team.append(0)\n",
    "\n",
    "        except TypeError as err:\n",
    "            # TypeError occurs when the team can't be found in the table\n",
    "            # Which means they were recently promoted from 1st division\n",
    "            top_5_team.append(0)\n",
    "        \n",
    "        # Calculating whether the opponent is a top 5 opponent\n",
    "        try:\n",
    "            if int(opponent_placement) < 6:\n",
    "                top_5_opponent.append(1)\n",
    "            else:\n",
    "                top_5_opponent.append(0)\n",
    "        \n",
    "        except TypeError as err:\n",
    "            # TypeError occurs when the opponent can't be found in the table\n",
    "            # Which means they were recently promoted from 1st division\n",
    "            top_5_opponent.append(0)\n",
    "        \n",
    "                      \n",
    "    if current_season == 2019:\n",
    "        team_placement = table_2018_df.loc[(table_2018_df[\"Squad\"] == team), \"Rk\"]\n",
    "        opponent_placement = table_2018_df.loc[(table_2018_df[\"Squad\"] == opponent), \"Rk\"]\n",
    "        \n",
    "        # Calculating whether the team is a top 5 team\n",
    "        try:\n",
    "            if int(team_placement) < 6:\n",
    "                top_5_team.append(1)\n",
    "            else:\n",
    "                top_5_team.append(0)\n",
    "        \n",
    "        except TypeError as err:\n",
    "            # TypeError occurs when the team can't be found in the table\n",
    "            # Which means they were recently promoted from 1st division\n",
    "            top_5_team.append(0)\n",
    "            \n",
    "        # Calculating whether the opponent is a top 5 opponent\n",
    "        try:\n",
    "            if int(opponent_placement) < 6:\n",
    "                top_5_opponent.append(1)\n",
    "            else:\n",
    "                top_5_opponent.append(0)\n",
    "        \n",
    "        except TypeError as err:\n",
    "            # TypeError occurs when the opponent can't be found in the table\n",
    "            # Which means they were recently promoted from 1st division\n",
    "            top_5_opponent.append(0)\n",
    "            \n",
    "    \n",
    "    if current_season == 2020:\n",
    "        team_placement = table_2019_df.loc[(table_2019_df[\"Squad\"] == team), \"Rk\"]\n",
    "        opponent_placement = table_2019_df.loc[(table_2019_df[\"Squad\"] == opponent), \"Rk\"]\n",
    "        \n",
    "        # Calculating whether the team is a top 5 team\n",
    "        try:\n",
    "            if int(team_placement) < 6:\n",
    "                top_5_team.append(1)\n",
    "            else:\n",
    "                top_5_team.append(0)\n",
    "        \n",
    "        except TypeError as err:\n",
    "            # TypeError occurs when the team can't be found in the table\n",
    "            # Which means they were recently promoted from 1st division\n",
    "            top_5_team.append(0)\n",
    "            \n",
    "        # Calculating whether the opponent is a top 5 opponent\n",
    "        try:\n",
    "            if int(opponent_placement) < 6:\n",
    "                top_5_opponent.append(1)\n",
    "            else:\n",
    "                top_5_opponent.append(0)\n",
    "        \n",
    "        except TypeError as err:\n",
    "            # TypeError occurs when the opponent can't be found in the table\n",
    "            # Which means they were recently promoted from 1st division\n",
    "            top_5_opponent.append(0)\n",
    "\n",
    "\n",
    "# Creating the two new features with the calculated values\n",
    "games_df[\"Top 5 Team\"] = top_5_team\n",
    "games_df[\"Top 5 Opponent\"] = top_5_opponent\n",
    "\n",
    "# Reordering the features again so that the predictor variables appear at the end\n",
    "games_df = games_df[[col for col in games_df.columns if col not in [\"y1: Team Score\", \"y2: Opponent Score\"]] + [\"y1: Team Score\", \"y2: Opponent Score\"]]\n",
    "\n",
    "# View the result\n",
    "display(games_df.head())\n",
    "display(games_df.tail())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "4466488c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Season</th>\n",
       "      <th>Team</th>\n",
       "      <th>Opponent</th>\n",
       "      <th>Venue</th>\n",
       "      <th>Game # in season</th>\n",
       "      <th>Avg. Historical Score Diff</th>\n",
       "      <th>Date</th>\n",
       "      <th>Team's goals per 90 min last season</th>\n",
       "      <th>Opponent's goals per 90 min last season</th>\n",
       "      <th>Team's Avg. Age</th>\n",
       "      <th>Opponent's Avg. Age</th>\n",
       "      <th>Top 5 Team</th>\n",
       "      <th>Top 5 Opponent</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2020</td>\n",
       "      <td>Trondheims-Ørn</td>\n",
       "      <td>Klepp</td>\n",
       "      <td>Home</td>\n",
       "      <td>1.0</td>\n",
       "      <td>-0.166667</td>\n",
       "      <td>2020-08-02</td>\n",
       "      <td>1.18</td>\n",
       "      <td>2.18</td>\n",
       "      <td>21.9</td>\n",
       "      <td>25.0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2020</td>\n",
       "      <td>Røa</td>\n",
       "      <td>Vålerenga</td>\n",
       "      <td>Away</td>\n",
       "      <td>1.0</td>\n",
       "      <td>-0.166667</td>\n",
       "      <td>2020-07-10</td>\n",
       "      <td>1.77</td>\n",
       "      <td>1.86</td>\n",
       "      <td>22.9</td>\n",
       "      <td>25.2</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2020</td>\n",
       "      <td>Trondheims-Ørn</td>\n",
       "      <td>Arna-Bjørnar</td>\n",
       "      <td>Home</td>\n",
       "      <td>2.0</td>\n",
       "      <td>-1.833333</td>\n",
       "      <td>2020-08-30</td>\n",
       "      <td>1.18</td>\n",
       "      <td>1.18</td>\n",
       "      <td>21.9</td>\n",
       "      <td>22.4</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2020</td>\n",
       "      <td>Vålerenga</td>\n",
       "      <td>Lyn</td>\n",
       "      <td>Home</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.250000</td>\n",
       "      <td>2020-08-26</td>\n",
       "      <td>1.86</td>\n",
       "      <td>1.05</td>\n",
       "      <td>25.2</td>\n",
       "      <td>20.1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2020</td>\n",
       "      <td>Arna-Bjørnar</td>\n",
       "      <td>Kolbotn</td>\n",
       "      <td>Home</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.166667</td>\n",
       "      <td>2020-09-12</td>\n",
       "      <td>1.18</td>\n",
       "      <td>1.59</td>\n",
       "      <td>22.4</td>\n",
       "      <td>21.8</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Season            Team      Opponent Venue  Game # in season  \\\n",
       "0    2020  Trondheims-Ørn         Klepp  Home               1.0   \n",
       "1    2020             Røa     Vålerenga  Away               1.0   \n",
       "2    2020  Trondheims-Ørn  Arna-Bjørnar  Home               2.0   \n",
       "3    2020       Vålerenga           Lyn  Home               1.0   \n",
       "4    2020    Arna-Bjørnar       Kolbotn  Home               1.0   \n",
       "\n",
       "   Avg. Historical Score Diff       Date  Team's goals per 90 min last season  \\\n",
       "0                   -0.166667 2020-08-02                                 1.18   \n",
       "1                   -0.166667 2020-07-10                                 1.77   \n",
       "2                   -1.833333 2020-08-30                                 1.18   \n",
       "3                    1.250000 2020-08-26                                 1.86   \n",
       "4                    1.166667 2020-09-12                                 1.18   \n",
       "\n",
       "   Opponent's goals per 90 min last season  Team's Avg. Age  \\\n",
       "0                                     2.18             21.9   \n",
       "1                                     1.86             22.9   \n",
       "2                                     1.18             21.9   \n",
       "3                                     1.05             25.2   \n",
       "4                                     1.59             22.4   \n",
       "\n",
       "   Opponent's Avg. Age  Top 5 Team  Top 5 Opponent  \n",
       "0                 25.0           0               1  \n",
       "1                 25.2           0               1  \n",
       "2                 22.4           0               0  \n",
       "3                 20.1           1               0  \n",
       "4                 21.8           0               0  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Season</th>\n",
       "      <th>Team</th>\n",
       "      <th>Opponent</th>\n",
       "      <th>Venue</th>\n",
       "      <th>Game # in season</th>\n",
       "      <th>Avg. Historical Score Diff</th>\n",
       "      <th>Date</th>\n",
       "      <th>Team's goals per 90 min last season</th>\n",
       "      <th>Opponent's goals per 90 min last season</th>\n",
       "      <th>Team's Avg. Age</th>\n",
       "      <th>Opponent's Avg. Age</th>\n",
       "      <th>Top 5 Team</th>\n",
       "      <th>Top 5 Opponent</th>\n",
       "      <th>y1: Team Score</th>\n",
       "      <th>y2: Opponent Score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>791</th>\n",
       "      <td>2019</td>\n",
       "      <td>Avaldsnes</td>\n",
       "      <td>Trondheims-Ørn</td>\n",
       "      <td>Home</td>\n",
       "      <td>22.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2019-11-16</td>\n",
       "      <td>1.090000</td>\n",
       "      <td>1.140000</td>\n",
       "      <td>24.400000</td>\n",
       "      <td>21.900000</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>792</th>\n",
       "      <td>2019</td>\n",
       "      <td>Lyn</td>\n",
       "      <td>IF Fløya</td>\n",
       "      <td>Away</td>\n",
       "      <td>23.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2019-11-24</td>\n",
       "      <td>1.180000</td>\n",
       "      <td>0.925741</td>\n",
       "      <td>20.100000</td>\n",
       "      <td>23.408333</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>793</th>\n",
       "      <td>2019</td>\n",
       "      <td>IF Fløya</td>\n",
       "      <td>Lyn</td>\n",
       "      <td>Home</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2019-11-24</td>\n",
       "      <td>0.925741</td>\n",
       "      <td>1.180000</td>\n",
       "      <td>23.408333</td>\n",
       "      <td>20.100000</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>5.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>794</th>\n",
       "      <td>2019</td>\n",
       "      <td>Lyn</td>\n",
       "      <td>IF Fløya</td>\n",
       "      <td>Home</td>\n",
       "      <td>24.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>2019-12-01</td>\n",
       "      <td>1.180000</td>\n",
       "      <td>0.925741</td>\n",
       "      <td>20.100000</td>\n",
       "      <td>23.408333</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>795</th>\n",
       "      <td>2019</td>\n",
       "      <td>IF Fløya</td>\n",
       "      <td>Lyn</td>\n",
       "      <td>Away</td>\n",
       "      <td>2.0</td>\n",
       "      <td>-5.0</td>\n",
       "      <td>2019-12-01</td>\n",
       "      <td>0.925741</td>\n",
       "      <td>1.180000</td>\n",
       "      <td>23.408333</td>\n",
       "      <td>20.100000</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     Season       Team        Opponent Venue  Game # in season  \\\n",
       "791    2019  Avaldsnes  Trondheims-Ørn  Home              22.0   \n",
       "792    2019        Lyn        IF Fløya  Away              23.0   \n",
       "793    2019   IF Fløya             Lyn  Home               1.0   \n",
       "794    2019        Lyn        IF Fløya  Home              24.0   \n",
       "795    2019   IF Fløya             Lyn  Away               2.0   \n",
       "\n",
       "     Avg. Historical Score Diff       Date  \\\n",
       "791                         1.0 2019-11-16   \n",
       "792                         0.0 2019-11-24   \n",
       "793                         0.0 2019-11-24   \n",
       "794                         5.0 2019-12-01   \n",
       "795                        -5.0 2019-12-01   \n",
       "\n",
       "     Team's goals per 90 min last season  \\\n",
       "791                             1.090000   \n",
       "792                             1.180000   \n",
       "793                             0.925741   \n",
       "794                             1.180000   \n",
       "795                             0.925741   \n",
       "\n",
       "     Opponent's goals per 90 min last season  Team's Avg. Age  \\\n",
       "791                                 1.140000        24.400000   \n",
       "792                                 0.925741        20.100000   \n",
       "793                                 1.180000        23.408333   \n",
       "794                                 0.925741        20.100000   \n",
       "795                                 1.180000        23.408333   \n",
       "\n",
       "     Opponent's Avg. Age  Top 5 Team  Top 5 Opponent  y1: Team Score  \\\n",
       "791            21.900000           0               0             0.0   \n",
       "792            23.408333           0               0             5.0   \n",
       "793            20.100000           0               0             0.0   \n",
       "794            23.408333           0               0             2.0   \n",
       "795            20.100000           0               0             1.0   \n",
       "\n",
       "     y2: Opponent Score  \n",
       "791                 1.0  \n",
       "792                 0.0  \n",
       "793                 5.0  \n",
       "794                 1.0  \n",
       "795                 2.0  "
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Separating the 2020 data from the games dataset\n",
    "# because I don't want to calculate descriptive statistics for it\n",
    "games_2020_df = games_df.loc[games_df[\"Season\"] == 2020].drop(columns=[\"y1: Team Score\", \"y2: Opponent Score\"])\n",
    "games_df = games_df.loc[games_df[\"Season\"] != 2020]\n",
    "display(games_2020_df.head())\n",
    "games_df.tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "805ebeac",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Season</th>\n",
       "      <th>Game # in season</th>\n",
       "      <th>Avg. Historical Score Diff</th>\n",
       "      <th>Team's goals per 90 min last season</th>\n",
       "      <th>Opponent's goals per 90 min last season</th>\n",
       "      <th>Team's Avg. Age</th>\n",
       "      <th>Opponent's Avg. Age</th>\n",
       "      <th>Top 5 Team</th>\n",
       "      <th>Top 5 Opponent</th>\n",
       "      <th>y1: Team Score</th>\n",
       "      <th>y2: Opponent Score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>796.000000</td>\n",
       "      <td>796.000000</td>\n",
       "      <td>7.960000e+02</td>\n",
       "      <td>796.000000</td>\n",
       "      <td>796.000000</td>\n",
       "      <td>796.000000</td>\n",
       "      <td>796.000000</td>\n",
       "      <td>796.000000</td>\n",
       "      <td>796.000000</td>\n",
       "      <td>796.000000</td>\n",
       "      <td>796.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>2018.005025</td>\n",
       "      <td>11.815327</td>\n",
       "      <td>2.231604e-18</td>\n",
       "      <td>1.638645</td>\n",
       "      <td>1.638645</td>\n",
       "      <td>23.373696</td>\n",
       "      <td>23.373696</td>\n",
       "      <td>0.276382</td>\n",
       "      <td>0.276382</td>\n",
       "      <td>1.616834</td>\n",
       "      <td>1.616834</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.818020</td>\n",
       "      <td>6.396525</td>\n",
       "      <td>1.665302e+00</td>\n",
       "      <td>0.488728</td>\n",
       "      <td>0.488728</td>\n",
       "      <td>1.756381</td>\n",
       "      <td>1.756381</td>\n",
       "      <td>0.447489</td>\n",
       "      <td>0.447489</td>\n",
       "      <td>1.515594</td>\n",
       "      <td>1.515594</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>2017.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-7.000000e+00</td>\n",
       "      <td>0.730000</td>\n",
       "      <td>0.730000</td>\n",
       "      <td>19.800000</td>\n",
       "      <td>19.800000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>2017.000000</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>-1.000000e+00</td>\n",
       "      <td>1.360000</td>\n",
       "      <td>1.360000</td>\n",
       "      <td>21.900000</td>\n",
       "      <td>21.900000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>2018.000000</td>\n",
       "      <td>12.000000</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>1.640152</td>\n",
       "      <td>1.640152</td>\n",
       "      <td>23.404167</td>\n",
       "      <td>23.404167</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>2019.000000</td>\n",
       "      <td>17.000000</td>\n",
       "      <td>1.000000e+00</td>\n",
       "      <td>1.730000</td>\n",
       "      <td>1.730000</td>\n",
       "      <td>25.100000</td>\n",
       "      <td>25.100000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>2019.000000</td>\n",
       "      <td>24.000000</td>\n",
       "      <td>7.000000e+00</td>\n",
       "      <td>3.090000</td>\n",
       "      <td>3.090000</td>\n",
       "      <td>26.400000</td>\n",
       "      <td>26.400000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>9.000000</td>\n",
       "      <td>9.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            Season  Game # in season  Avg. Historical Score Diff  \\\n",
       "count   796.000000        796.000000                7.960000e+02   \n",
       "mean   2018.005025         11.815327                2.231604e-18   \n",
       "std       0.818020          6.396525                1.665302e+00   \n",
       "min    2017.000000          1.000000               -7.000000e+00   \n",
       "25%    2017.000000          6.000000               -1.000000e+00   \n",
       "50%    2018.000000         12.000000                0.000000e+00   \n",
       "75%    2019.000000         17.000000                1.000000e+00   \n",
       "max    2019.000000         24.000000                7.000000e+00   \n",
       "\n",
       "       Team's goals per 90 min last season  \\\n",
       "count                           796.000000   \n",
       "mean                              1.638645   \n",
       "std                               0.488728   \n",
       "min                               0.730000   \n",
       "25%                               1.360000   \n",
       "50%                               1.640152   \n",
       "75%                               1.730000   \n",
       "max                               3.090000   \n",
       "\n",
       "       Opponent's goals per 90 min last season  Team's Avg. Age  \\\n",
       "count                               796.000000       796.000000   \n",
       "mean                                  1.638645        23.373696   \n",
       "std                                   0.488728         1.756381   \n",
       "min                                   0.730000        19.800000   \n",
       "25%                                   1.360000        21.900000   \n",
       "50%                                   1.640152        23.404167   \n",
       "75%                                   1.730000        25.100000   \n",
       "max                                   3.090000        26.400000   \n",
       "\n",
       "       Opponent's Avg. Age  Top 5 Team  Top 5 Opponent  y1: Team Score  \\\n",
       "count           796.000000  796.000000      796.000000      796.000000   \n",
       "mean             23.373696    0.276382        0.276382        1.616834   \n",
       "std               1.756381    0.447489        0.447489        1.515594   \n",
       "min              19.800000    0.000000        0.000000        0.000000   \n",
       "25%              21.900000    0.000000        0.000000        0.000000   \n",
       "50%              23.404167    0.000000        0.000000        1.000000   \n",
       "75%              25.100000    1.000000        1.000000        2.000000   \n",
       "max              26.400000    1.000000        1.000000        9.000000   \n",
       "\n",
       "       y2: Opponent Score  \n",
       "count          796.000000  \n",
       "mean             1.616834  \n",
       "std              1.515594  \n",
       "min              0.000000  \n",
       "25%              0.000000  \n",
       "50%              1.000000  \n",
       "75%              2.000000  \n",
       "max              9.000000  "
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "games_df.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8ed7690a",
   "metadata": {},
   "source": [
    "### <a id=ohe></a>3.10) One Hot Encoding categorical features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "a94837ac",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Season</th>\n",
       "      <th>Team</th>\n",
       "      <th>Opponent</th>\n",
       "      <th>Venue</th>\n",
       "      <th>Game # in season</th>\n",
       "      <th>Avg. Historical Score Diff</th>\n",
       "      <th>Date</th>\n",
       "      <th>Team's goals per 90 min last season</th>\n",
       "      <th>Opponent's goals per 90 min last season</th>\n",
       "      <th>Team's Avg. Age</th>\n",
       "      <th>Opponent's Avg. Age</th>\n",
       "      <th>Top 5 Team</th>\n",
       "      <th>Top 5 Opponent</th>\n",
       "      <th>y1: Team Score</th>\n",
       "      <th>y2: Opponent Score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>179</th>\n",
       "      <td>2020</td>\n",
       "      <td>Arna-Bjørnar</td>\n",
       "      <td>Sandviken</td>\n",
       "      <td>Away</td>\n",
       "      <td>17.0</td>\n",
       "      <td>-0.571429</td>\n",
       "      <td>2020-09-27</td>\n",
       "      <td>1.18</td>\n",
       "      <td>1.95</td>\n",
       "      <td>22.4</td>\n",
       "      <td>25.3</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>180</th>\n",
       "      <td>2020</td>\n",
       "      <td>Røa</td>\n",
       "      <td>LSK Kvinner</td>\n",
       "      <td>Home</td>\n",
       "      <td>18.0</td>\n",
       "      <td>-2.000000</td>\n",
       "      <td>2020-07-17</td>\n",
       "      <td>1.77</td>\n",
       "      <td>2.27</td>\n",
       "      <td>22.9</td>\n",
       "      <td>25.3</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>181</th>\n",
       "      <td>2020</td>\n",
       "      <td>Kolbotn</td>\n",
       "      <td>LSK Kvinner</td>\n",
       "      <td>Home</td>\n",
       "      <td>20.0</td>\n",
       "      <td>-1.714286</td>\n",
       "      <td>2020-11-07</td>\n",
       "      <td>1.59</td>\n",
       "      <td>2.27</td>\n",
       "      <td>21.8</td>\n",
       "      <td>25.3</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>182</th>\n",
       "      <td>2020</td>\n",
       "      <td>Arna-Bjørnar</td>\n",
       "      <td>Røa</td>\n",
       "      <td>Away</td>\n",
       "      <td>18.0</td>\n",
       "      <td>-0.571429</td>\n",
       "      <td>2020-07-04</td>\n",
       "      <td>1.18</td>\n",
       "      <td>1.77</td>\n",
       "      <td>22.4</td>\n",
       "      <td>22.9</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>183</th>\n",
       "      <td>2020</td>\n",
       "      <td>Sandviken</td>\n",
       "      <td>Lyn</td>\n",
       "      <td>Away</td>\n",
       "      <td>18.0</td>\n",
       "      <td>1.800000</td>\n",
       "      <td>2020-09-12</td>\n",
       "      <td>1.95</td>\n",
       "      <td>1.05</td>\n",
       "      <td>25.3</td>\n",
       "      <td>20.1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     Season          Team     Opponent Venue  Game # in season  \\\n",
       "179    2020  Arna-Bjørnar    Sandviken  Away              17.0   \n",
       "180    2020           Røa  LSK Kvinner  Home              18.0   \n",
       "181    2020       Kolbotn  LSK Kvinner  Home              20.0   \n",
       "182    2020  Arna-Bjørnar          Røa  Away              18.0   \n",
       "183    2020     Sandviken          Lyn  Away              18.0   \n",
       "\n",
       "     Avg. Historical Score Diff       Date  \\\n",
       "179                   -0.571429 2020-09-27   \n",
       "180                   -2.000000 2020-07-17   \n",
       "181                   -1.714286 2020-11-07   \n",
       "182                   -0.571429 2020-07-04   \n",
       "183                    1.800000 2020-09-12   \n",
       "\n",
       "     Team's goals per 90 min last season  \\\n",
       "179                                 1.18   \n",
       "180                                 1.77   \n",
       "181                                 1.59   \n",
       "182                                 1.18   \n",
       "183                                 1.95   \n",
       "\n",
       "     Opponent's goals per 90 min last season  Team's Avg. Age  \\\n",
       "179                                     1.95             22.4   \n",
       "180                                     2.27             22.9   \n",
       "181                                     2.27             21.8   \n",
       "182                                     1.77             22.4   \n",
       "183                                     1.05             25.3   \n",
       "\n",
       "     Opponent's Avg. Age  Top 5 Team  Top 5 Opponent  y1: Team Score  \\\n",
       "179                 25.3           0               1             NaN   \n",
       "180                 25.3           0               1             NaN   \n",
       "181                 25.3           0               1             NaN   \n",
       "182                 22.9           0               0             NaN   \n",
       "183                 20.1           1               0             NaN   \n",
       "\n",
       "     y2: Opponent Score  \n",
       "179                 NaN  \n",
       "180                 NaN  \n",
       "181                 NaN  \n",
       "182                 NaN  \n",
       "183                 NaN  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                   FEATURE                     DTYPE\n",
      "------------------------------------------|--------------|\n",
      "Season                                              int64\n",
      "Game # in season                                  float64\n",
      "Avg. Historical Score Diff                        float64\n",
      "Date                                       datetime64[ns]\n",
      "Team's goals per 90 min last season               float64\n",
      "Opponent's goals per 90 min last season           float64\n",
      "Team's Avg. Age                                   float64\n",
      "Opponent's Avg. Age                               float64\n",
      "Top 5 Team                                          int64\n",
      "Top 5 Opponent                                      int64\n",
      "y1: Team Score                                    float64\n",
      "y2: Opponent Score                                float64\n",
      "Team_Arna-Bjørnar                                   uint8\n",
      "Team_Avaldsnes                                      uint8\n",
      "Team_Fart                                           uint8\n",
      "Team_Grand Bodø                                     uint8\n",
      "Team_IF Fløya                                       uint8\n",
      "Team_Klepp                                          uint8\n",
      "Team_Kolbotn                                        uint8\n",
      "Team_LSK Kvinner                                    uint8\n",
      "Team_Lyn                                            uint8\n",
      "Team_Medkila                                        uint8\n",
      "Team_Røa                                            uint8\n",
      "Team_Sandviken                                      uint8\n",
      "Team_Stabæk                                         uint8\n",
      "Team_Trondheims-Ørn                                 uint8\n",
      "Team_Vålerenga                                      uint8\n",
      "Opponent_Arna-Bjørnar                               uint8\n",
      "Opponent_Avaldsnes                                  uint8\n",
      "Opponent_Fart                                       uint8\n",
      "Opponent_Grand Bodø                                 uint8\n",
      "Opponent_IF Fløya                                   uint8\n",
      "Opponent_Klepp                                      uint8\n",
      "Opponent_Kolbotn                                    uint8\n",
      "Opponent_LSK Kvinner                                uint8\n",
      "Opponent_Lyn                                        uint8\n",
      "Opponent_Medkila                                    uint8\n",
      "Opponent_Røa                                        uint8\n",
      "Opponent_Sandviken                                  uint8\n",
      "Opponent_Stabæk                                     uint8\n",
      "Opponent_Trondheims-Ørn                             uint8\n",
      "Opponent_Vålerenga                                  uint8\n",
      "Venue_Away                                          uint8\n",
      "Venue_Home                                          uint8\n",
      "dtype: object\n"
     ]
    }
   ],
   "source": [
    "# Re-joining the games data\n",
    "games_df = pd.concat([games_df, games_2020_df], axis=0)\n",
    "display(games_df.tail())\n",
    "\n",
    "# One Hot Encoding of categorical variables\n",
    "games_df = pd.get_dummies(games_df)\n",
    "print(f\"                   FEATURE                     DTYPE\")\n",
    "print(\"------------------------------------------|--------------|\")\n",
    "print(games_df.dtypes)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "88c67175",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Reordering columns\n",
    "games_df = games_df[[\"Date\", \"Season\", 'Team_Arna-Bjørnar', 'Team_Avaldsnes',\n",
    "       'Team_Fart', 'Team_Grand Bodø', 'Team_IF Fløya', 'Team_Klepp',\n",
    "       'Team_Kolbotn', 'Team_LSK Kvinner', 'Team_Lyn', 'Team_Medkila',\n",
    "       'Team_Røa', 'Team_Sandviken', 'Team_Stabæk', 'Team_Trondheims-Ørn',\n",
    "       'Team_Vålerenga', 'Opponent_Arna-Bjørnar', 'Opponent_Avaldsnes',\n",
    "       'Opponent_Fart', 'Opponent_Grand Bodø', 'Opponent_IF Fløya',\n",
    "       'Opponent_Klepp', 'Opponent_Kolbotn', 'Opponent_LSK Kvinner',\n",
    "       'Opponent_Lyn', 'Opponent_Medkila', 'Opponent_Røa',\n",
    "       'Opponent_Sandviken', 'Opponent_Stabæk', 'Opponent_Trondheims-Ørn',\n",
    "       'Opponent_Vålerenga', 'Venue_Away', 'Venue_Home', \"Game # in season\",\n",
    "       \"Avg. Historical Score Diff\", \n",
    "       \"Team's goals per 90 min last season\",\n",
    "       \"Opponent's goals per 90 min last season\", \"Team's Avg. Age\",\n",
    "       \"Opponent's Avg. Age\", \"Top 5 Team\", \"Top 5 Opponent\", \"y1: Team Score\",\n",
    "       \"y2: Opponent Score\"]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "1a73956c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>ID</th>\n",
       "      <th>Season</th>\n",
       "      <th>Team_Arna-Bjørnar</th>\n",
       "      <th>Team_Avaldsnes</th>\n",
       "      <th>Team_Fart</th>\n",
       "      <th>Team_Grand Bodø</th>\n",
       "      <th>Team_IF Fløya</th>\n",
       "      <th>Team_Klepp</th>\n",
       "      <th>Team_Kolbotn</th>\n",
       "      <th>Team_LSK Kvinner</th>\n",
       "      <th>Team_Lyn</th>\n",
       "      <th>Team_Medkila</th>\n",
       "      <th>Team_Røa</th>\n",
       "      <th>Team_Sandviken</th>\n",
       "      <th>Team_Stabæk</th>\n",
       "      <th>Team_Trondheims-Ørn</th>\n",
       "      <th>Team_Vålerenga</th>\n",
       "      <th>Opponent_Arna-Bjørnar</th>\n",
       "      <th>Opponent_Avaldsnes</th>\n",
       "      <th>Opponent_Fart</th>\n",
       "      <th>Opponent_Grand Bodø</th>\n",
       "      <th>Opponent_IF Fløya</th>\n",
       "      <th>Opponent_Klepp</th>\n",
       "      <th>Opponent_Kolbotn</th>\n",
       "      <th>Opponent_LSK Kvinner</th>\n",
       "      <th>Opponent_Lyn</th>\n",
       "      <th>Opponent_Medkila</th>\n",
       "      <th>Opponent_Røa</th>\n",
       "      <th>Opponent_Sandviken</th>\n",
       "      <th>Opponent_Stabæk</th>\n",
       "      <th>Opponent_Trondheims-Ørn</th>\n",
       "      <th>Opponent_Vålerenga</th>\n",
       "      <th>Venue_Away</th>\n",
       "      <th>Venue_Home</th>\n",
       "      <th>Game # in season</th>\n",
       "      <th>Avg. Historical Score Diff</th>\n",
       "      <th>Team's goals per 90 min last season</th>\n",
       "      <th>Opponent's goals per 90 min last season</th>\n",
       "      <th>Team's Avg. Age</th>\n",
       "      <th>Opponent's Avg. Age</th>\n",
       "      <th>Top 5 Team</th>\n",
       "      <th>Top 5 Opponent</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>179</th>\n",
       "      <td>179</td>\n",
       "      <td>2020</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>17.0</td>\n",
       "      <td>-0.571429</td>\n",
       "      <td>1.18</td>\n",
       "      <td>1.95</td>\n",
       "      <td>22.4</td>\n",
       "      <td>25.3</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>180</th>\n",
       "      <td>180</td>\n",
       "      <td>2020</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>18.0</td>\n",
       "      <td>-2.000000</td>\n",
       "      <td>1.77</td>\n",
       "      <td>2.27</td>\n",
       "      <td>22.9</td>\n",
       "      <td>25.3</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>181</th>\n",
       "      <td>181</td>\n",
       "      <td>2020</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>20.0</td>\n",
       "      <td>-1.714286</td>\n",
       "      <td>1.59</td>\n",
       "      <td>2.27</td>\n",
       "      <td>21.8</td>\n",
       "      <td>25.3</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>182</th>\n",
       "      <td>182</td>\n",
       "      <td>2020</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>18.0</td>\n",
       "      <td>-0.571429</td>\n",
       "      <td>1.18</td>\n",
       "      <td>1.77</td>\n",
       "      <td>22.4</td>\n",
       "      <td>22.9</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>183</th>\n",
       "      <td>183</td>\n",
       "      <td>2020</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>18.0</td>\n",
       "      <td>1.800000</td>\n",
       "      <td>1.95</td>\n",
       "      <td>1.05</td>\n",
       "      <td>25.3</td>\n",
       "      <td>20.1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      ID  Season  Team_Arna-Bjørnar  Team_Avaldsnes  Team_Fart  \\\n",
       "179  179    2020                  1               0          0   \n",
       "180  180    2020                  0               0          0   \n",
       "181  181    2020                  0               0          0   \n",
       "182  182    2020                  1               0          0   \n",
       "183  183    2020                  0               0          0   \n",
       "\n",
       "     Team_Grand Bodø  Team_IF Fløya  Team_Klepp  Team_Kolbotn  \\\n",
       "179                0              0           0             0   \n",
       "180                0              0           0             0   \n",
       "181                0              0           0             1   \n",
       "182                0              0           0             0   \n",
       "183                0              0           0             0   \n",
       "\n",
       "     Team_LSK Kvinner  Team_Lyn  Team_Medkila  Team_Røa  Team_Sandviken  \\\n",
       "179                 0         0             0         0               0   \n",
       "180                 0         0             0         1               0   \n",
       "181                 0         0             0         0               0   \n",
       "182                 0         0             0         0               0   \n",
       "183                 0         0             0         0               1   \n",
       "\n",
       "     Team_Stabæk  Team_Trondheims-Ørn  Team_Vålerenga  Opponent_Arna-Bjørnar  \\\n",
       "179            0                    0               0                      0   \n",
       "180            0                    0               0                      0   \n",
       "181            0                    0               0                      0   \n",
       "182            0                    0               0                      0   \n",
       "183            0                    0               0                      0   \n",
       "\n",
       "     Opponent_Avaldsnes  Opponent_Fart  Opponent_Grand Bodø  \\\n",
       "179                   0              0                    0   \n",
       "180                   0              0                    0   \n",
       "181                   0              0                    0   \n",
       "182                   0              0                    0   \n",
       "183                   0              0                    0   \n",
       "\n",
       "     Opponent_IF Fløya  Opponent_Klepp  Opponent_Kolbotn  \\\n",
       "179                  0               0                 0   \n",
       "180                  0               0                 0   \n",
       "181                  0               0                 0   \n",
       "182                  0               0                 0   \n",
       "183                  0               0                 0   \n",
       "\n",
       "     Opponent_LSK Kvinner  Opponent_Lyn  Opponent_Medkila  Opponent_Røa  \\\n",
       "179                     0             0                 0             0   \n",
       "180                     1             0                 0             0   \n",
       "181                     1             0                 0             0   \n",
       "182                     0             0                 0             1   \n",
       "183                     0             1                 0             0   \n",
       "\n",
       "     Opponent_Sandviken  Opponent_Stabæk  Opponent_Trondheims-Ørn  \\\n",
       "179                   1                0                        0   \n",
       "180                   0                0                        0   \n",
       "181                   0                0                        0   \n",
       "182                   0                0                        0   \n",
       "183                   0                0                        0   \n",
       "\n",
       "     Opponent_Vålerenga  Venue_Away  Venue_Home  Game # in season  \\\n",
       "179                   0           1           0              17.0   \n",
       "180                   0           0           1              18.0   \n",
       "181                   0           0           1              20.0   \n",
       "182                   0           1           0              18.0   \n",
       "183                   0           1           0              18.0   \n",
       "\n",
       "     Avg. Historical Score Diff  Team's goals per 90 min last season  \\\n",
       "179                   -0.571429                                 1.18   \n",
       "180                   -2.000000                                 1.77   \n",
       "181                   -1.714286                                 1.59   \n",
       "182                   -0.571429                                 1.18   \n",
       "183                    1.800000                                 1.95   \n",
       "\n",
       "     Opponent's goals per 90 min last season  Team's Avg. Age  \\\n",
       "179                                     1.95             22.4   \n",
       "180                                     2.27             22.9   \n",
       "181                                     2.27             21.8   \n",
       "182                                     1.77             22.4   \n",
       "183                                     1.05             25.3   \n",
       "\n",
       "     Opponent's Avg. Age  Top 5 Team  Top 5 Opponent  \n",
       "179                 25.3           0               1  \n",
       "180                 25.3           0               1  \n",
       "181                 25.3           0               1  \n",
       "182                 22.9           0               0  \n",
       "183                 20.1           1               0  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Separating the 2020 data, again\n",
    "games_2020_df = games_df.loc[games_df[\"Season\"] == 2020].drop(columns=[\"y1: Team Score\", \"y2: Opponent Score\", \n",
    "                                                                       \"Date\"])\n",
    "games_df = games_df.loc[games_df[\"Season\"] != 2020]\n",
    "games_df.tail()\n",
    "\n",
    "# Reattach game IDs to the 2020 game data\n",
    "games_2020_df = pd.concat([game_ids, games_2020_df], axis=1)\n",
    "display(games_2020_df.tail())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b7381129",
   "metadata": {},
   "source": [
    "[*back to top*](#top)\n",
    "## <a id=\"saving\"></a>4) Saving the cleaned data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "1bb0f866",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Cleaned data looks like this:\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Date</th>\n",
       "      <th>Season</th>\n",
       "      <th>Team_Arna-Bjørnar</th>\n",
       "      <th>Team_Avaldsnes</th>\n",
       "      <th>Team_Fart</th>\n",
       "      <th>Team_Grand Bodø</th>\n",
       "      <th>Team_IF Fløya</th>\n",
       "      <th>Team_Klepp</th>\n",
       "      <th>Team_Kolbotn</th>\n",
       "      <th>Team_LSK Kvinner</th>\n",
       "      <th>Team_Lyn</th>\n",
       "      <th>Team_Medkila</th>\n",
       "      <th>Team_Røa</th>\n",
       "      <th>Team_Sandviken</th>\n",
       "      <th>Team_Stabæk</th>\n",
       "      <th>Team_Trondheims-Ørn</th>\n",
       "      <th>Team_Vålerenga</th>\n",
       "      <th>Opponent_Arna-Bjørnar</th>\n",
       "      <th>Opponent_Avaldsnes</th>\n",
       "      <th>Opponent_Fart</th>\n",
       "      <th>Opponent_Grand Bodø</th>\n",
       "      <th>Opponent_IF Fløya</th>\n",
       "      <th>Opponent_Klepp</th>\n",
       "      <th>Opponent_Kolbotn</th>\n",
       "      <th>Opponent_LSK Kvinner</th>\n",
       "      <th>Opponent_Lyn</th>\n",
       "      <th>Opponent_Medkila</th>\n",
       "      <th>Opponent_Røa</th>\n",
       "      <th>Opponent_Sandviken</th>\n",
       "      <th>Opponent_Stabæk</th>\n",
       "      <th>Opponent_Trondheims-Ørn</th>\n",
       "      <th>Opponent_Vålerenga</th>\n",
       "      <th>Venue_Away</th>\n",
       "      <th>Venue_Home</th>\n",
       "      <th>Game # in season</th>\n",
       "      <th>Avg. Historical Score Diff</th>\n",
       "      <th>Team's goals per 90 min last season</th>\n",
       "      <th>Opponent's goals per 90 min last season</th>\n",
       "      <th>Team's Avg. Age</th>\n",
       "      <th>Opponent's Avg. Age</th>\n",
       "      <th>Top 5 Team</th>\n",
       "      <th>Top 5 Opponent</th>\n",
       "      <th>y1: Team Score</th>\n",
       "      <th>y2: Opponent Score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>2017</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>-0.166667</td>\n",
       "      <td>1.640152</td>\n",
       "      <td>1.640152</td>\n",
       "      <td>21.9</td>\n",
       "      <td>25.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>2017</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>-0.166667</td>\n",
       "      <td>1.640152</td>\n",
       "      <td>1.640152</td>\n",
       "      <td>22.9</td>\n",
       "      <td>25.2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>2017</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>-1.833333</td>\n",
       "      <td>1.640152</td>\n",
       "      <td>1.640152</td>\n",
       "      <td>21.9</td>\n",
       "      <td>22.4</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>2017</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.250000</td>\n",
       "      <td>1.640152</td>\n",
       "      <td>1.640152</td>\n",
       "      <td>25.2</td>\n",
       "      <td>20.1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>6.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2017-04-17</td>\n",
       "      <td>2017</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.166667</td>\n",
       "      <td>1.640152</td>\n",
       "      <td>1.640152</td>\n",
       "      <td>22.4</td>\n",
       "      <td>21.8</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        Date  Season  Team_Arna-Bjørnar  Team_Avaldsnes  Team_Fart  \\\n",
       "0 2017-04-17    2017                  0               0          0   \n",
       "1 2017-04-17    2017                  1               0          0   \n",
       "2 2017-04-17    2017                  0               0          0   \n",
       "3 2017-04-17    2017                  0               0          0   \n",
       "4 2017-04-17    2017                  0               0          0   \n",
       "\n",
       "   Team_Grand Bodø  Team_IF Fløya  Team_Klepp  Team_Kolbotn  Team_LSK Kvinner  \\\n",
       "0                0              0           0             0                 0   \n",
       "1                0              0           0             0                 0   \n",
       "2                0              0           0             0                 0   \n",
       "3                0              0           1             0                 0   \n",
       "4                0              0           0             1                 0   \n",
       "\n",
       "   Team_Lyn  Team_Medkila  Team_Røa  Team_Sandviken  Team_Stabæk  \\\n",
       "0         0             1         0               0            0   \n",
       "1         0             0         0               0            0   \n",
       "2         0             0         0               0            1   \n",
       "3         0             0         0               0            0   \n",
       "4         0             0         0               0            0   \n",
       "\n",
       "   Team_Trondheims-Ørn  Team_Vålerenga  Opponent_Arna-Bjørnar  \\\n",
       "0                    0               0                      0   \n",
       "1                    0               0                      0   \n",
       "2                    0               0                      0   \n",
       "3                    0               0                      0   \n",
       "4                    0               0                      0   \n",
       "\n",
       "   Opponent_Avaldsnes  Opponent_Fart  Opponent_Grand Bodø  Opponent_IF Fløya  \\\n",
       "0                   0              0                    0                  0   \n",
       "1                   0              0                    1                  0   \n",
       "2                   0              0                    0                  0   \n",
       "3                   0              0                    0                  0   \n",
       "4                   0              0                    0                  0   \n",
       "\n",
       "   Opponent_Klepp  Opponent_Kolbotn  Opponent_LSK Kvinner  Opponent_Lyn  \\\n",
       "0               0                 0                     0             0   \n",
       "1               0                 0                     0             0   \n",
       "2               0                 0                     0             0   \n",
       "3               0                 0                     1             0   \n",
       "4               0                 0                     0             0   \n",
       "\n",
       "   Opponent_Medkila  Opponent_Røa  Opponent_Sandviken  Opponent_Stabæk  \\\n",
       "0                 0             0                   1                0   \n",
       "1                 0             0                   0                0   \n",
       "2                 0             1                   0                0   \n",
       "3                 0             0                   0                0   \n",
       "4                 0             0                   0                0   \n",
       "\n",
       "   Opponent_Trondheims-Ørn  Opponent_Vålerenga  Venue_Away  Venue_Home  \\\n",
       "0                        0                   0           0           1   \n",
       "1                        0                   0           1           0   \n",
       "2                        0                   0           1           0   \n",
       "3                        0                   0           1           0   \n",
       "4                        1                   0           1           0   \n",
       "\n",
       "   Game # in season  Avg. Historical Score Diff  \\\n",
       "0               1.0                   -0.166667   \n",
       "1               1.0                   -0.166667   \n",
       "2               2.0                   -1.833333   \n",
       "3               1.0                    1.250000   \n",
       "4               1.0                    1.166667   \n",
       "\n",
       "   Team's goals per 90 min last season  \\\n",
       "0                             1.640152   \n",
       "1                             1.640152   \n",
       "2                             1.640152   \n",
       "3                             1.640152   \n",
       "4                             1.640152   \n",
       "\n",
       "   Opponent's goals per 90 min last season  Team's Avg. Age  \\\n",
       "0                                 1.640152             21.9   \n",
       "1                                 1.640152             22.9   \n",
       "2                                 1.640152             21.9   \n",
       "3                                 1.640152             25.2   \n",
       "4                                 1.640152             22.4   \n",
       "\n",
       "   Opponent's Avg. Age  Top 5 Team  Top 5 Opponent  y1: Team Score  \\\n",
       "0                 25.0           0               0             1.0   \n",
       "1                 25.2           0               0             2.0   \n",
       "2                 22.4           0               0             2.0   \n",
       "3                 20.1           0               0             1.0   \n",
       "4                 21.8           0               0             2.0   \n",
       "\n",
       "   y2: Opponent Score  \n",
       "0                 1.0  \n",
       "1                 2.0  \n",
       "2                 1.0  \n",
       "3                 6.0  \n",
       "4                 3.0  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "print(\"Cleaned data looks like this:\")\n",
    "display(games_df.head())\n",
    "\n",
    "# Saving the cleaned data as .csv\n",
    "games_df.to_csv(\"Prepared Data/games_prepared.csv\", index=False, encoding=\"UTF-8\")  # for training and evaluating ML models\n",
    "games_2020_df.to_csv(\"Prepared Data/games_2020_prepared.csv\", index=False, encoding=\"UTF-8\")  # for predicting on 2020 data\n",
    "teams_2019_df.to_csv(\"Prepared Data/2019_team-stats_prepared.csv\")  # for preprocessing website input\n",
    "table_2019_df.to_csv(\"Prepared Data/2019_table_prepared.csv\", index=False)\n",
    "\n",
    "# Saving the cleaned data as .html\n",
    "games_df.to_html(\"Prepared Data/games_prepared.html\", index=False, encoding=\"UTF-8\")\n",
    "games_2020_df.to_html(\"Prepared Data/games_2020_prepared.html\", index=False, encoding=\"UTF-8\")"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}