Frequently Asked Questions
Updated throughout the competition. Don't see your question? Post it in the Q&A session on April 27.
Who is eligible to participate?
Any currently enrolled student of the Gies College of Business at the University of Illinois Urbana-Champaign.
What is the team size?
Exactly two students per team. No solo registrations, no teams of three or more.
What is the submission deadline?
Friday, May 1, 2026 at 11:59 PM Central Time. A 2-hour grace period follows until 2:00 AM CT May 2, after which late submissions are flagged on the leaderboard.
What do we submit?
Four components: (1) a completed prediction Excel file, (2) your code / model files, (3) a written methodology PDF, and (4) a PowerPoint presentation.
How is scoring done?
An automated scorer computes weighted MAPE (Mean Absolute Percentage Error) on your predictions against Horizon Hobby's held-back actuals for three parts. The latest version of your prediction file is what gets scored.
Is the leaderboard ranking the same as winning the competition?
No. The leaderboard ranks forecast accuracy on the three held-out Old SKU parts using weighted MAPE. But the three finalists who present to Horizon Hobby are selected by a panel that also weighs your methodology and rationale in the PDF + code, and your business insight for the New SKUs in the presentation. A strong write-up with modest accuracy can beat a top-accuracy submission with weak analysis. See the How you're judged section on the landing page for the full rubric.
Do the New SKU predictions count if they aren't on the leaderboard?
Yes — just not on the leaderboard. The New SKU predictions (the Dirt Late Model parts) are reviewed directly by Horizon Hobby for finalist selection, because those forecasts are the ones they'll actually use for inventory and launch planning. Treat them as the business deliverable, not an optional add-on. You can use your wMAPE on the held-out Old SKU parts as a validation signal to tune your approach for the New SKUs.