Duke season-to-date: 2-0 ATS, 51.5% average bet share, 36.0% money share.
| Metric | Value |
|---|---|
| ATS record (tracked games) | 2-0 |
| Average bet % | 51.5% |
| Average money % | 36.0% |
| Games as public favorite (60%+ bets) | 1 |
| ATS record as public favorite | 1-0 |
| Fade-the-public record (vs them as public fav) | 0-1 |
Duke don't have a strong public-vs-money pattern this season. Across the 2 games we've tracked, the average bet percentage on Duke is 51.5% and the average money percentage is 36.0%. Those are middle-of-the-pack numbers — bettors are split, dollars follow the bet count, and the line generally reflects the matchup rather than the public lean.
Sample size is a real consideration: 2 games is not enough to call a season-long pattern. The "heavy fade target" or "heavy public favorite" labels we apply to other teams require at least 5 games of data and a meaningful directional lean — neither applies here yet.
Watch the per-game splits below. Even balanced teams have individual games with meaningful divergence — that's where the daily roundups come in. See our methodology →.
Our advanced model rates every NCAAF game by expected value, using thousands of historical splits as the backbone. This week's highest-EV picks are reserved for subscribers.
Want a free look? See this week's biggest splits →
| Date | Matchup | Result | Duke bet % | Duke money % |
|---|---|---|---|---|
| 2026-09-12 | Duke @ Illinois | Covered 31-27 (+6) | 38.0% | 34.0% |
| 2026-09-05 | Tulane @ Duke | Covered 17-3 (-8.5) | 65.0% | 38.0% |
0-1 ATS this season — the mirror of their public-favorite record, because the fade covers exactly when the public side doesn't. It's shown as a record rather than a recommendation; a handful of games is not a pattern.
From publicly reported sportsbook handle, captured on a sub-hourly cadence for every FBS game from the moment a line is posted. The team figures use each game's closing snapshot — the last read before kickoff — so every number reflects what a bettor could actually have known.
A game counts only when it has a final score, a closing spread and a published split. An FCS opponent the source never priced, or a game whose splits were not captured before kickoff, is left out rather than guessed at.
When 60% or more of spread tickets land on Duke, they're 1-0 ATS this season. Books shade college lines toward the side they know the public wants, and on a board full of mismatches that shade is often the whole story.
Duke are 2-0 against the spread in the games where we have a closing split. That's the number that matters for a bettor — a team can win every Saturday by three touchdowns and still lose money for the people laying four.
How we track public bets and money — see our methodology →
Last updated: September 15, 2026 at 2:52 AM UTC
Every game on the slate, ranked by public-vs-money divergence. Splits-based top opportunities, updated continuously. No signup, no paywall.
See all of this week's splits →