First Win / Free research
How accurate are college football betting odds?
A five-season test of moneylines, spreads and totals around ten minutes before kickoff, across the validated matched-game universe.
The quick take
The probabilities held up better than the prices.
College football pregame prices looked broadly compatible with what happened on the field, but this study was not precise enough to certify one-percentage-point accuracy. Every calibration interval included zero, while all three selected strategies lost at the prices offered and had return intervals entirely below zero.
Follow the evidence
Calibration and return
Points are estimates. Bars are 95% source-season/calendar-week block bootstrap intervals.
The numbers
| Market and selection | Games | Expected | Actual | Gap | 95% calendar-week interval | Return | Return interval | Pushes |
|---|---|---|---|---|---|---|---|---|
| Moneyline favorites | 3,859 | 72.42% | 72.38% | -0.04pp | [-1.22, 1.11]pp | -4.25% | [-5.99, -2.57]% | 0 |
| Spread favorites | 3,913 | 50.07% | 49.49% | -0.58pp | [-2.02, 0.77]pp | -5.29% | [-8.03, -2.69]% | 76 |
| Totals overs | 3,923 | 50.06% | 49.67% | -0.40pp | [-2.02, 1.26]pp | -5.10% | [-8.20, -1.93]% | 39 |
Expected win rates remove each book's margin first. Returns use the observed quoted prices. Pushes return the stake and are excluded from conditional win-rate and decided-stake return denominators.
What it means
A market can estimate probabilities reasonably well while still offering losing prices. Calibration asks whether 60% propositions win about 60% of the time. Return asks what happened after paying the sportsbook's margin.
The negative returns need restraint. They do not turn underdogs or unders into demonstrated strategies. Complementary sides used the same games and paid their own margin.
The catch
This is a retrospective benchmark. The 2021-2025 outcomes were visible during data validation, so none is an untouched confirmation sample. These are the latest eligible quotes around ten minutes before scheduled kickoff, not guaranteed final closing prices or guaranteed fills.
The intervals still failed the sealed one-point precision target. Only 4,031 of 4,742 manifest games had confirmed provider matches, and the source does not separate bowls and conference championships from other games. A future frozen season and broader identity reconciliation are needed.
Methodology, version and provenance
Ten historically available books. Per-book proportional de-vigging followed by an equal-book average. Spreads and totals use the modal quoted line. Research QA, deliverable QA and deterministic replay passed.
First Win Pro
Build a benchmark that states what its coverage misses.
The companion note turns this result into a research standard: matching market, line, price and pregame window, plus the uncertainty hurdle an apparent edge should clear.
Read the Pro companionContinue reading
Get new sports-market research when we publish
New research, straight to your inbox. No picks, no promotions. Our public studies stay free to read without an account.