Sports Info Solutions (SIS) is hosting a Football Analytics Challenge to raise money for the Boys and Girls Clubs of America. All donations are appreciated and go to charity.
Learn more about the competition here.
Visit this link to register for the Football Analytics Challenge; a minimum $1 donation is required for entry.
Data is available in the Data folder, with a Data Dictionary file to explain the use of the files and some of the columns.
Participants can collaborate and form teams of up to five members to create a joint submission. Each team or individual participant is limited to submitting one official entry.
After registering, participants will retrieve contest dataset through GitHub and must submit their work through the same GitHub contest platform to be considered eligible.
Participants authorized to use any additional data that is both free and publicly available. Participants must identify and cite data that goes beyond supplied contest data.
Start Date: June 28, 2021 Entry Deadline (Final Submission Deadline): July 25, 2021 11:59 PM ET Preliminary Judging Results: July 30, 2021 Final Judging Panel: August 4, 2021 (live video call)
Submissions are due by 11:59pm ET on Sunday, July 25. All submissions are final. See Submissions folder for instructions on how to submit your work.
All eligible entrants will have their work reviewed in a preliminary judging round conducted by the Sports Info Solutions Research & Development Department. Each submission will be judged by the evaluation criteria below. Three finalists will be selected to advance to the final judging round. Finalists must be available to participate in the live judging round on August 4, 2021 via a live Zoom call.
This competition is being held with two separate tracks: a General Track that emphasizes evaluating football as a whole, and a Gaming Track that focuses on edges that can be gained in fantasy and/or betting contexts. Participants are welcome to submit entries to both tracks, provided that they do not make more than one final submission on a given track.
Which route combinations were most popular in the NFL in 2020? Of these route combinations, which perform best against each coverage type?
How much of an effect should quarterbacks and wide receivers have on a point spread? Responses must provide estimates for Patrick Mahomes and Julio Jones and also include other prominent players from the 2020 season as examples.
You will submit the answers to these questions in the form of a presentation (e.g. PowerPoint or Google Slides, R Markdown), providing any additional information about the analysis that led you to your conclusions. The document provided should include, but is not limited to, the items below. Their weight in the grading process is also included.
• Final answers (15%) — The answers to the above question(s), enumerated explicitly so they are easy to identify. While there isn’t going to be a “right” answer to any of these questions, it is crucial that: a) you provide clear and concise answers to the question(s), and b) the answer(s) you give are based on results found over the course of this project.
• Assumptions (10%) — The prompts are intentionally a little bit broad so that participants have a chance to work through things in different ways. If appropriate, make sure to clarify any assumptions you might be making about information that wasn’t specified as part of the prompt, and explain why you settled on those assumptions. For example: How are you defining the different route combinations you’re considering?
• Evaluation Method (50%) — What methods did you employ to answer the question of interest? This includes any analytical techniques used, but also will take into account the quality of the thought process while evaluating the question(s).
• Limitations and Future Analysis (10%) — We only provided you a slice of data to work with, and besides, no analysis is complete and final. What additional analysis might you have performed to answer the prompt(s) if you had more expansive resources, and what resources would those be?
• Presentation / Clarity of Communication (15%) — In any research context, it’s critical to be able to communicate the results of an analysis clearly and concisely. Convey the answers to the above question(s) as well as your methods in a clear and direct way that engages the audience.
• Other Data Sources — We are in a golden age of football analytics. Feel free to use other free, publicly available resources to aid in your analysis. You must provide details on any resources that you used outside the provided data, and it must be clear from your discussion what that data was used for. Of course, this competition is centered around the data provided, so don’t lean too heavily on other resources for your key conclusions.
The finalists will be notified of their progress via email by July 30th, 2021. If a finalist does not respond to the notification attempt within 48 hours from the first notification attempt, then an alternate finalist will be selected from among the eligible entrant pool.
To be deemed eligible, participants must submit a monetary donation of no less than $1 to the affiliated donation collection on GoFundMe and must register via the competition website. There are no geographic restrictions, but work must be submitted in English. Full-time employees of Sports Info Solutions are not eligible to participate.
This is a charity event with all proceeds donated to the Boys and Girls Clubs of America. Finalists will receive the unique opportunity to present their work to a panel of industry professionals. Finalist presentations may be promoted via SIS social media and blog. Winner will be invited to appear on the Off The Charts podcast with Matt Manocherian. Winner will also receive a free one-year subscription to the SIS DataHub Pro.
The contest dataset provided by Sports Info Solutions shall not be used for any purposes beyond this competition. Participants may not distribute, sell, transmit, brand, publish, copy, sublicense, reproduce, or transfer any of the contest data at any time during or after the competition without the prior written permission of SIS.
Participants shall retain all Intellectual Property rights to their Intellectual Property, including publishing the results of their research on any platform they wish. For avoidance of doubt, Participants may not share the underlying data under any circumstances. Sports Info Solutions retains the right to use Participant research and results on their website and in media in perpetuity at no cost, including work and models throughout future NFL Seasons and Offseasons.
This competition is a skills-based competition to promote and further the field of data science and raise money for charity. You submit a monetary donation of no less than $1 to the affiliated donation collection and must register via the competition website to enter. Your competition submission must conform to the requirements stated in this document. Your submission will be scored based on the evaluation criteria provided in this document.
Questions? contact [email protected]