analysis_nba: STAT 405 Final

Table of Contents

• General info
• Technologies
• Setup

General info

This project is a fast, condensed program to query NBA box score data, advanced box score data, and game reprots such as individual game summaries and officals' data. A single year takes roughly 6-8 hours to fully process.

For the year entered, three csv formmated files are filled:

• box_score_plus: Per game data on each player for every game in the season
• game_summaries: Game-by-game summary data on the teams that played, the broadcasting network, the game time, the game date, and the attendence
• total_officials: The names of the officials that refereed

box_score_plus is formatted as follows:

Basic Info	Box Score	Adv. Box Score	Misc. Box Score
Header 'GAME_ID' 'TEAM_ID' 'TEAM_ABBREVIATION' 'TEAM_CITY' 'PLAYER_ID' 'PLAYER_NAME' 'START_POSITION' 'COMMENT'	Header Definition 'MIN' Minutes per game 'FGM' Field Goals made 'FGA' Field Goals attempted 'FG_PCT' Field Goal % 'FG3A' 3 point attempts 'FG3M' 3 point makes 'FG3_PCT' 3 point % 'FTM' Free Throws made 'FTA' Free Throws attempted 'FT_PCT' Free Throw % 'DREB' Defensive Rebounds 'OREB' Offensive Rebounds 'REB' Rebounds 'AST' Assists 'STL' Steals 'BLK' Blocks 'TO' Turnovers 'PF' Personal Fouls 'PTS' Points 'PLUS_MINUS' Plus Minus	Header Definition 'OPP_OREB_PCT' Opponent's Offensive Rebound % 'OPP_TOV_PCT' Opponent's Turnover % 'OREB_PCT' Player's Offensive Rebound % 'TM_TOV_PCT' Team Turnover % 'BLKA' Player's # Block Attempts 'OPP_PTS_2ND_CHANCE' Opponent's 2nd chance points scored	Header Definition 'OPP_PTS_FB' Opponent Fast Break Points 'OPP_PTS_OFF_TOV' Opponent Points Off Turnovers 'OPP_PTS_PAINT' Opponent Points in the Paint 'PFD' Player's Personal fouls drawn 'PTS_2ND_CHANCE' Player's 2nd chance points scored 'PTS_FB' Player's Fast Break Points 'PTS_OFF_TOV' Player's Points Off Turnovers 'PTS_PAINT' Player's Points in the Paint

game_summaries and total_officials are formatted as follows:

Officals	Game Summaries/th>
Header 'GAMECODE' 'GAME_ID' 'GAME_DATE' 'OFFICIAL_ID' 'FIRST_NAME' 'LAST_NAME' 'JERSEY_NUM'	Header 'GAME_DATE_EST' 'GAME_ID' 'GAME_STATUS_ID' 'GAMECODE' 'HOME_TEAM_ID' 'VISITOR_TEAM_ID' 'SEASON' 'LIVE_PERIOD' 'NATL_TV_BROADCASTER_ABBREVIATION' 'LIVE_PERIOD_TIME_BCAST' 'WH_STATUS' 'GAME_DATE' 'ATTENDANCE'

Technologies

Project is created with:

• Python 3.6, R
• py_ball: For NBA API Calls
• matplotlib: Dependency for the py_ball package
• pandas: For dataframe formatting
• requests_cache: For API calls, SQLite storing
• dyplr: For dataframe concatenation and processing

Setup

The following instructions are for Mac OS.

Follow the instructions and run the commands in Terminal:

Step 0: cd into a folder that you want to store the code/files generated

$ cd /Users/{Insert username}/Documents/{Whatever path you want to use}

Step 0: Clone the repository

$ git clone https://github.com/awx1/analysis_nba.git

Step 1: Install pip

$ curl https://bootstrap.pypa.io/get-pip.py -o get-pip.py
$ python get-pip.py

Step 0: Install virtualenv

$ python3 -m pip install --user virtualenv
$ pip3 install virtualenv

Step 3: Access virtual env

$ python3 -m venv nba
$ source env/bin/activate

Step 4: Install packages

$ pip3 install cairosvg==2.2.1
$ pip3 install matplotlib==3.2.1
$ pip3 install pkginfo==1.5.0.1
$ pip3 install requests==2.20.0
$ pip3 install py_ball
$ pip3 install matplotlib
$ pip3 install pandas
$ pip3 install requests_cache

Step 5: Install homebrew

$ /bin/bash -c "$(curl -fsSL https://github.com/raw/Homebrew/install/master/install.sh)"

Step 6: Install cairo

$ brew install cairo

Step 7: Place this line of code at the bottom of final_copy.py (Specify the YEAR)

### Change the year here
get_year_boxscores({YEAR_HERE})

Step 9: Run the program using the command

python3 final_copy.py

Step 10: To concatenate the csv's using the R script, change the path to the correct path to run