Across 1,687 NFL games from 2020 to 2025, the consensus favorite won outright 67.0% of the time and covered the spread 49.0% of decided games. The straight-up rate rises steadily with a shorter price; the cover rate stays close to half.
Today's prices are on the NFL odds board.
The headline
Those two numbers say different things. Favorites win most games — that is what makes them favorites, and the price reflects it. Against the spread they sit close to half, which is what a functioning market looks like: the handicap is set so that neither side is the obvious one.
By price band
The interesting question is whether the straight-up win rate tracks the price. It does.
| Favorite's moneyline | Games | Won outright | Games with a spread | Covered (of decided) | Pushed |
|---|---|---|---|---|---|
| −100 to −129 | 166 | 55.4% | 166 | 52.7% | 0.6% |
| −130 to −159 | 327 | 57.8% | 327 | 48.6% | 4.3% |
| −160 to −199 | 312 | 61.5% | 312 | 49.3% | 3.9% |
| −200 to −249 | 203 | 63.0% | 203 | 44.5% | 0.5% |
| −250 to −349 | 287 | 72.8% | 287 | 49.6% | 1.7% |
| −350 or shorter | 392 | 81.6% | 392 | 49.2% | 2.0% |
Favorite identified from the consensus moneyline at the last pre-kickoff snapshot we hold, requiring at least three sportsbooks on both sides. Cover rate excludes pushes from the denominator. Regular season and playoffs, 2020–2025.
Read across the win column and the ladder is orderly: the shorter the price, the more often the favorite won. That is the market doing its job, not a finding about beatable prices.
The cover column behaves differently. It stays near 50% across the bands, and the individual band samples are small enough that the differences between them are within ordinary sampling noise. We are not claiming any band covers more often than another.
Sample sizes, stated plainly
Every band above carries its own n. Some are thin — a band with under a hundred games supports a rough statement about direction and nothing sharper. The pooled row is the one with real weight: 1,687 games straight up and 1,687 against the spread.
Does the reference point change the answer?
We re-ran the whole study against the first pre-game snapshot we hold for each game instead of the last. The pooled figures move very little, which is what you would expect if the result is a property of the market rather than of our sampling:
| Reference snapshot | Games | Favorite won outright |
|---|---|---|
| Last before kickoff | 1,687 | 66.98% |
| First we hold | 1,682 | 65.58% |
Both rows use the same games; only the snapshot the favorite is identified from changes.
What this is not
This is a description of what already happened. It is not a forecast, and it carries no suggestion about what to do with any future game. Nebula Insights publishes no picks and no recommended wagers — the methodology page explains why, and the public track record shows every projection we do publish, logged before the game.
How this was measured
- Universe: NFL games from 2020 to 2025 (regular season and playoffs) that we could match from our odds archive to a final result. 1,693 of 1,693 games matched.
- Favorite = the side with the shorter consensus moneyline at the reference snapshot, requiring at least three sportsbooks quoting both sides at that instant.
- Consensus uses a discrete median across books, so every price and line quoted is one a book actually offered.
- The shortest band has no upper bound, and the archive contains a small number of suspended-market quotes. Of 1,687 favorites, 37 carry a consensus price shorter than −1000 and 5 shorter than −2000. They are left in rather than trimmed, and counted here so the bottom band can be read honestly.
- Cover computed against the same-snapshot consensus spread for the favorite; pushes are reported separately and excluded from the cover denominator.
- Our snapshots sit on a daily grid, so “last before kickoff” means the last day we observed, not the closing line. We hold no kickoff-minute price and publish nothing that would require one.
- Produced by
repo_check/seo/f31_lake_stats.pyon a read-only database session. - Odds archive current to 2026-07-28. Figures refresh when the source scripts are re-run.
Next steps
Frequently asked questions
How often do NFL favorites win?
67.0% of the time across 1,687 games from 2020 to 2025, using the consensus moneyline favorite at the last pre-kickoff snapshot we hold.
How often do NFL favorites cover the spread?
49.0% of decided games, from 1,687 games with a usable consensus spread. 2.4% pushed.
Do heavier favorites cover more often?
Not measurably in our sample. Straight-up win rates rise clearly with a shorter price, but cover rates stay near half and the band-level differences are within sampling noise.
Is this a prediction about future games?
No. It is a record of what already happened. Nebula Insights publishes no picks and no recommended wagers.
Related: NFL odds board · NFL key numbers · track record · methodology
Every figure on this page is a historical measurement of what already happened, produced by a committed, re-runnable script over our own odds archive. Past frequencies are not forecasts and are not betting advice: Nebula Insights publishes no picks, no selections and no recommended wagers. 18+ only.