DATA-DRIVEN SPORTS INSIGHTS

Model the Game. Find the Edge.

Independent predictive models for baseball, basketball, and football. Every forecast is locked before the game, graded afterward, and kept on the record.

View today’s predictions →Explore Premium

Data-driven models

Thousands of simulations power our probabilistic projections.

Verified performance

Every pick is graded in public. Nothing is hidden.

Market awareness

We compare our numbers to the market to find real edges.

Research first

Continuous testing and validation keep our models sharp.

MethodologyHow the NCAAF model works

A team-strength composite anchored on opponent-adjusted efficiency ratings, with in-season box-score form blended in on a schedule that starts conservative and lets current play take over as the sample builds.

What it reads

SP+ efficiency ratings (offense, defense, special teams), the structural backbone
In-season box-score form, recency-weighted, held to a minority of the signal
Schedule quality, valued by the tier of opponent beaten or lost to
Venue, rest, bye weeks and short weeks
Game-day weather on the totals side

What it produces

A percentile-ranked strength score per FBS team
Projected spread, compressed at the extremes so a mismatch does not print an absurd number
Projected total from a points-per-play model
Win probability, and a full-season simulation behind the playoff picture

What we publish

Spread against the market, the primary product
Totals, monitored rather than promoted
Rankings, playoff odds and CFP field projection

Not claimed

The box-score half of the signal is not opponent-adjusted, which is why it is held down early in the season. The totals number is blended with the market, so it is not an independent read the way the spread is.

See the NCAAF record →