nba-props

Different settings in src/main.ipynb for DraftKings or FanDuel (SITE), Classic or Single Game contests (MODE).
Input provided contest files from DFS sites in data/ as current-{site}.csv; if Single Game, add -sg before .csv in file name.
- Only manual step required from user besides toggling desired settings.
Alerts to let you know what players have had props added since last run-through.
- Initial run through will spit out large list of names
Calculates fpts according to site rules, hence requirement to toggle setting.
- Uses provided salaries to determine best allocation of salary for players as FPTS / $1,000, fpts/$.
Factors in implied probabilities of player props for separate value of "Expected Fantasy Points", e_fpts.
- Similarly uses this value with salary to determine best allocation, e_fpts/$
Uses industy standard value of 5 x (salary/$1,000) as another metric, heavily correlates with previous two values
Organizes players by site provided positions and teams.
Calculates distribution of players by team in order to be cognizant of over representation of some teams.
- Later games and teams with injuries tend to have props updated later in day

To install this is just like installing any other GitHub repository.
I have noticed I've received lots of clones, and I assume it is individuals in the DFS / Sports Betting space, so I figured I'd add some directions for people not used to dealing with Python super heavily.
On Linux or MacOS: Installation is done from the Command Line so open Terminal or equivalent.
- (windows users, I'm working on it)
(optional) If want to install in location that is not home directory:

$ cd path/to/target/directory

$ git clone https://github.com/jackdegen/nba-props

(optional) If you want to create a virtual environment, .venv; if you prefer a different name, repalce .venv with it:
- Note: if you do not do this step and you do not use Python, you will have to manually install pandas, requests, BeautifulSoup, and all other necessary packages on your system with pip. This may require dealing with different system dependencies
- I strongly suggest following these steps, even if you are not entirely sure what they are doing.
- They are just installing the external libraries required to run the code only in this directory, not on your entire machine.

$ python3 -m venv .venv
$ source .venv/bin/activate
(.venv) $ pip install -r requirements.txt

If you have JupyterLab/JupyterNotebook, I suggest running the code in that as it is easier on the eyes and more interactive if you are familiar with Jupyter.
If not, check back here in a few days as I configure the code to work in the Python shell of the command line.

$ jupyter-lab

This should have opened up a Jupyter Notebook in your browser.
Go to the navigation menu, and open src/main.py.
Toggle settings in first cell for SITE and MODE.
Run the cells either one by one or with fast-forward button, just be wary of last cell.
Need to be careful with the cell containing the following: handler.constant_scrape():
- Running an infinite loop to update data and add any new players.
- However, it will need to be manually stopped in order to make notebook interactive again.
- Doing so will trigger a KeyboardInterrupt error, that is not an actual error it is a result of shutting down the loop.
- All data will have been saved and updated behind the scenes, will not be any output since it runs quietly in background.
A new file will have been created in data/ containing the info for the NBA slate that day for whichever site you specified.
You can read this in with pd.read_csv() or simply access it using handler to get the data and assign it to a new variable if you would like to further interact with the dataset in a notebook.
The PropHandler class already has some built in functionality for basic data analysis, so feel free to consult the source code in src/prophandler/prophandler.py.

Contributing

Pull requests are welcome. For major changes, please open an issue first to discuss what you would like to change.

Please make sure to update tests as appropriate.