Inspiration
Our inspiration for this project was our interest in NFL and that we all play Fantasy. This led us to be very interested in the rookies that come into the league every year. We wanted to use the mass amounts of data tracked during seasons in order to make predictions about future rookie NFL players.
What it does
Our analysis shows undervalued rookies in the past and tries to make a prediction into the upcoming 2026 NFL season. It also takes a deep dive into team performance and strengths/weaknesses per player or position.
How we built it
We built his analysis leveraging many Python libraries, that including Pandas, Matplotlib, Scikit-learn, Seaborn, and many more. To find our data, Kaggle and football-reference were vital sources of information.
Challenges we ran into
Some challenges that we ran into during development was finding/sourcing meaningful data that was up to date. We found that a lot of the good data was blocked behind paywalls and much of the free data was older information we did not find useful as part of this observation.
Accomplishments that we're proud of
Some accomplishments that we are especially proud of is delivering and preparing these metrics and findings in an easy to read and timely manner. Additionally, Alex and Ismail had never done Data Science, so this was a good first step into understanding how the field worked.
What we learned
We learned a plethora of skills, especially those surrounding the data science scope. Specifically, sourcing, cleaning, and making graphs on that data were new points that we all had to learn. This also resulted in a more rewarding result, where we found observations that we could use in the future in our fantasy leagues for years to come.
What's next for NFL Combine & Rookie Insights
This was a great first step in taking a deeper look at the NFL and how rookies are placed in the draft. Given more time or a continuation of resources, we would be able to further our findings and get more realistic, useful statistics that we can use for next season. Overall, this was a rewarding and interesting project that gave us some great insights and predictions for the 2026 season of the NFL.
Built With
- chatgpt
- gemini
- jupyter
- matplotlib
- pandas
- python
- scikit-learn
- seaborn
Log in or sign up for Devpost to join the conversation.