Saturday rates every team in points, then plays each matchup thousands of times to get a range of outcomes instead of one score.
Home strength minus away strength, plus home field. Each team's number starts in the spring from last year, recruiting, the portal and coaching, then moves with every week of results. A game is never priced with anything that happened after kickoff.
Saturday rated Ohio State about 11.1 points stronger than Penn State on a neutral field. Add 3.3 for playing in Columbus and you get Ohio State by 14.4.
A worked example from week 10 of 2025, not this week's projection.
The formula is public: what goes into a rating.
“Ohio State by 14.4” is the middle of a range, not a call on the final score. Play the game a thousand times and it lands everywhere from a rout to an upset. The win chance is the share of the curve on the winning side of zero.
The curve is widest in August, when Saturday has watched none of this season, and narrows as results come in.
Every call is graded against the final score and the market's opening line, hits and misses alike. See the report card.
One input comes from the market: preseason win totals help set August ratings, then fade to nothing. No game is priced off a line.
Every setting and every constant, enough to rebuild the model.