For The Love of Basketball

The Gameplan

The idea for this project came from a shared interest in learning about how the coronavirus pandemic has shaken and shifted our cultural landscape. We found that, at this point in time, analyzing data for the NBA and WNBA presented a useful microcosm for how the world of professional sports and other large events.

Having mostly-complete metrics for complete seasons from both gender-based leagues, gave us a more comprehensive picture of performance before the pandemic and during it, at both the player and team levels. These insights allowed us to build predictive models for what the future of the game will hold as indoor stadiums remain largely empty across the United States.

- We decided to develop a machine learning-based algorithm to predict NBA and WNBA game results. To achieve this goal, we built a machine learning model to make predictions for both NBA and WNBA games – that is, predicting the probability of each team winning an NBA game, as well as presenting the win points behind the predictions.

- We decided to apply machine learning on predicting NBA and WNBA game results. NBA is one of the most popular sports league in the world, so it is not surprising that NBA fans would be eager to know who will win in the NBA season. If accurate predictions on NBA and WNBA game results could be done by utilizing machine learning, it will help create more excitement and engagement for NBA & WNBA fans all over the world.

- For the last NBA & WNBA season, our model obtained an overall accuracy of --%. To further optimize the results, we are eager to try ?....Recurrent Neural Network (RNN).

Next Steps

The insights presented here have lots of real-world applications. We also found more opportunities for going deeper into the data if given more time. Here are some of our plans for going forward:

- Improve our accurancy and enhance insight by gaining historical data from the beginning of both leagues.

- Develop a new machine learning model that will take player stats and predict who the player is.

- Extend our concept to other leagues such as the MLB and NFL.

- Explore the limits of how accurately we can predict NBA & WNBA games, and to understand more about machine learning’s capability and limitations on making predictions. We can also further explore the opportunities of applying machine learning on more dynamic situations and create more business values with technology.