Five seasons of college football modeling across every P4 game, built from drive efficiency data and schedule-adjusted team ratings.
Each component addresses a distinct source of predictive information that the college football market consistently misprices.
College football runs on drives, not plays. We evaluate each team's points-per-drive and opponent-adjusted drive success rate, the same framework that powers our NFL model, calibrated for college scoring environments.
A team going 8-0 in the Sun Belt is not the same as going 5-3 in the SEC. We apply iterative SOS adjustment so ratings reflect true quality, not record. Three iterations to convergence, the same approach used in college basketball's RPI replacement.
61% of our lines move toward our number after Monday release, the highest CLV rate of any Whizard spread model. We price before the market catches up. Early release gives subscribers access to the number before sharp money moves it.
From raw drive data to a betting edge. Five stages, no black boxes.
We measure how efficiently each team converts drives into points, adjusted for opponent quality. Yards per play lies; points per drive tells the truth.
A team averaging 8.2 points per drive against a strong schedule is genuinely elite. A team averaging 7.5 against cupcakes is not. The adjustment separates the two before we build any projection.
An iterative schedule-strength adjustment ensures a team's rating reflects the quality of opponents they've faced. Three iterations to convergence, the same principle used in the best college football ranking systems.
A team like Appalachian State with gaudy raw numbers against a weak schedule sees its rating cut sharply. A team like Georgia playing an elite SEC slate gains rating even with a moderate win-loss record.
College home field advantages are the most extreme in sports. A team playing in front of 100,000 fans in a hostile night game has a measurable edge. We model it per program, not a uniform NFL-style adjustment.
Alabama, Michigan, and Ohio State have historically elite home field edges built on decades of crowd atmosphere data. Smaller programs in smaller venues carry proportionally smaller home advantages.
The college football schedule creates predictable mispricing: revenge games, rivalry inflated lines, trap games before rivalry week. We identify these patterns before the market does.
Public bettors inflate lines in rivalry games and look-ahead spots. Our model prices the actual drive efficiency matchup independently of narrative, then flags the divergence when the market drifts from the number.
When our model line diverges 3+ points from the opening, before the market moves. That is our play. 61% CLV confirms we're consistently ahead of the public money.
The goal is not volume. A 20-play week at 61% CLV is better than a 60-play week at 54%. We only release plays that clear our minimum edge threshold against the opening number.
Five seasons of NCAAF spread plays, logged against closing line value across all P4 conferences. 61% CLV rate is the clearest signal of consistent edge.
NCAAF spreads and totals across all P4 games, released Monday before lines move.
Picks for the full week's P4 slate are released every Monday morning, ahead of typical line movement. Early access gives you the best opportunity to get on the opening number before sharp money arrives.
Every release includes model outputs for both markets. Each play shows our projected line, the edge versus the opening, and the market price at release. SOS ratings are refreshed every week as new game data arrives.
Schedule-strength ratings are updated every week as the season develops. Early-season ratings are inherently noisier; subscribers are notified of confidence tiers so they can size plays accordingly.
Early-season plays (Weeks 1-3) carry lower confidence as SOS data is limited. Confidence increases substantially from Week 4 onward as opponent ratings stabilize.
All picks land directly in the Whizard subscriber Discord. Each post includes the model output card, edge breakdown, situational flags, and unit recommendation.
NCAAF picks return for the 2026 season. Check back in August. In the meantime, explore our active models.
NCAAF picks return Fall 2026. Check back in August for the preseason model reset, updated SOS ratings, and Week 1 releases.