NFL preseason gets a bad reputation among bettors. The starters play a series and walk off, the depth charts are scrambled, and the games themselves often look like organized chaos. But the betting market for preseason games has a structure of its own — and a small group of professional bettors quietly attack it every August. This guide walks through the NFL preseason betting patterns that show up year after year, why sharp money moves preseason lines despite the noise, and where the public money keeps getting it wrong.
Why preseason markets are different
Three structural facts drive everything else:
- Starter snaps are tightly capped. Most starters play 8-15 snaps in week 1, then 1-2 series in week 2, and sit week 3 entirely. Quarterback usage is the most variable input on any given Saturday; Josh Allen sitting after one drive is a different market than a rookie QB playing a full half.
- Coaches have specific evaluation goals. The "depth chart battle" framing is real. A team competing for an interior OL spot will run the same group for 3 quarters to evaluate them.
- Public bets the brand. Cowboys, Eagles, 49ers, Chiefs, and Bills see public money pour in regardless of who is actually playing. That distorts the line away from the matchup-specific reality.
The sharp pattern: small bets, late movement
The professional approach to preseason is different from regular-season betting in two ways. First, position sizes are smaller. A bettor who plays $5,000 a side on regular-season games might play $500 on a preseason game. Second, the timing matters more. Most preseason sharp action lands in the final 24-48 hours before kickoff, after coaches release their snap-count plans and beat reporters confirm which starters will see the field.
That late-week movement is the cleanest signal in the preseason market. If a line opens at +3 on Tuesday and moves to -1 by Friday afternoon with no public money on that side, you are watching sharp action. The opposite is also true: a line that drifts away from a popular team despite heavy public ticket count means the books have set a number against the public and the sharps are agreeing. The regular-season version of that read is explained in sharp money vs public money.
Where to find edge each week
Week 1: Backup vs starter mismatches
In week 1, most teams play their established starters for 1-2 series and the backups for the rest. The matchups that matter are 2nd-string vs 2nd-string and 3rd-string vs 3rd-string. Teams with deeper rosters (well-drafted teams, contenders building depth) systematically outperform thin rosters in week 1, even with starters not finishing the game.
Week 2: The depth-chart battle
Week 2 is the highest-volume preseason week in terms of meaningful reps. Coaches give their bubble players extended runs to make a roster decision. Lines move on news of which positional battles are still open vs decided. A team that has clearly settled their QB1 depth chart will play the QB1 less; a team with an active QB1/QB2 battle will play both meaningfully.
Week 3: The starter-rest week
Week 3 (or whatever the second-to-last preseason week is) is the most asymmetric week of the calendar. Teams that have decided their starters sit them. Teams that have not decided their starters play them deep into the third quarter. The mismatch in playing time creates spreads that look strange against power ratings — and the spreads are usually right.
Public bias by team
Some teams attract preseason public money disproportionately every year:
- Cowboys — by far the heaviest preseason public team. Lines on Dallas are typically shaded a half point against the public bias.
- Eagles, 49ers, Chiefs, Bills — strong national followings, strong public ticket counts, especially when Jalen Hurts, Patrick Mahomes, or Josh Allen are listed as possible starters.
- Recently drafted hype QBs — any team with a rookie quarterback the public wants to see plays inflated public ticket counts.
The pattern: when the line on a heavy-public team has not moved despite 70%+ of tickets, the books have planted a flag. Fading the public in those spots over 3+ years tends to produce a small positive ROI, especially in week 1.
Totals: where the bigger edge lives
Preseason totals are softer than spreads because the books spend less time on them. Two patterns repeat:
- Unders in week 3. Starters sit, third-string offenses struggle to move the ball, and totals priced off regular-season expectations come in. Week 3 unders historically cash at a noticeable rate.
- Overs when both teams have depth issues. Two thin teams playing 3rd-string defenses produces high-variance scoring drives. The under looks safe on paper and gets blown up by a 65-yard run.
A worked example with concrete numbers
Suppose a week 2 preseason matchup opens at:
Cowboys -2.5, total 39.5
By Friday morning, beat reporters confirm Dallas is sitting starters by the second quarter while their opponent will play their starters into the third. The line drifts to:
Cowboys -1, total 41.5
What happened: sharps faded the public Cowboys money on the spread, and the total moved up because the opponent's starter usage means more first-half scoring drives. If you saw that movement on Tuesday and waited for confirmation Friday, you have a clear read: the opening line was wrong, the sharp side has been confirmed, and the matchup-specific factor (starter playing time) is the dominant input.
Common mistakes
- Treating preseason like regular season. Power ratings barely matter. Snap-count plans matter. Depth-chart battles matter.
- Overweighting recent regular-season form. A team that finished strong last December is not the same team in August with 30 roster turnover.
- Betting big. Position sizing matters in preseason because the variance is enormous. A starter twisting an ankle on snap 8 changes the entire game. Use bankroll rules, not regular-season confidence.
- Ignoring rest-week dynamics. Teams playing their second preseason game in 5 days play differently than teams on standard 7-day rest.
How to read the line move
Three useful filters for any preseason line:
- Direction vs public ticket count. If the line moves toward the side with fewer tickets, that is sharp money. If it moves with the public, the book is rebalancing exposure.
- Time of move. Late-week moves (Friday onward) are higher signal than early-week. Early moves are reactive to reports; late moves are reactive to confirmed lineup info.
- Cross-book agreement. When all major books move the same direction within an hour, that is a market-wide adjustment, not a single book responding to one big bet.
Where models help in preseason
A model that learned the regular season — point spreads, power ratings, win-loss — is useless in August because none of what actually decides these games (who plays, how long, which depth-chart battles are live) is in front of it. The fix is a model built specifically for preseason that bets off snap counts and starter usage, or — more practically — using your regular-season read only as a rough starting point and adjusting by hand for who's playing. Our NFL picks page generally avoids preseason because the edge is thin compared to the regular-season grind.
Bottom line
NFL preseason is a real market for the disciplined bettor: small position sizes, late-week timing, and a focus on snap-count plans rather than power ratings. Sharp money lives in the final 24-48 hours when starter usage gets confirmed. Public bias on the Cowboys, Eagles, and rookie-QB teams is a recurring fade pattern. Totals are softer than spreads, and week 3 unders are the cleanest specific angle.
If you are using preseason as a warmup for the regular-season grind, treat it like a learning lab — track your line-reading accuracy, build the habit of waiting for late info, and size small. For ongoing picks, see the bet-tracking guide for logging process and the model builder when the regular-season schedule drops.
Take this into a real preseason workflow
The right way to run preseason as a learning lab is to wire the workflow into Shark Snip rather than guessing in a spreadsheet. Open the Workshop with a preseason-specific topic preset and build a model that makes the call for you off the inputs that actually move these games — snap counts by quarter, first-team reps in the red zone, two-minute work — pointed at whichever market you care about: preseason spread, total over-under, or a starter prop. Then test it on the last five preseasons to see if it would have beaten the number. Start a fresh model if you want full control over how it weighs each input, or copy a published preseason model from the creator marketplace as a starting point. Whichever path you pick, the discipline is the same: small positions, late-week timing, snap-count thesis, and an honest closing-line-value log. Push your weekly results to the creator leaderboard to see whether your preseason reads actually beat the close, or take a no-bankroll squad into the NFL auto-battler if you want to feel how a preseason-tuned read survives a simulated regular season before any real money moves in September.
Bet responsibly — set limits, never chase losses.
Price examples and pass rules
Use names as evidence, not decoration. The useful SEO win is that Patrick Mahomes, Josh Allen, Jalen Hurts, Ja'Marr Chase and Bijan Robinson and Chiefs, Bills, Eagles, Cowboys and 49ers appear inside decisions, thresholds, and internal links instead of being dumped into a keyword list.
- Spread example: if Chiefs-Broncos opens Chiefs -3.5 and your fair number is -2.8, +3.5 is the bet, +3 is a pass, and the moneyline needs roughly +155 or better before it replaces the spread.
- Total example: if a Bills outdoor total opens 46.5 and wind moves from 8 mph to 21 mph, an under projection at 42.8 still needs a playable number; under 45 or better is different from chasing 43.5.
- Futures example: Bengals AFC North +280 is 26.3% before hold. If your fair number is 30%, stake modestly, track portfolio correlation, and avoid stacking every Burrow, Chase, and Higgins bet into the same thesis.
- CLV rule: a good write-up is not enough. Track whether the spread, total, prop, or futures price closed better than your entry before grading the process.
Use closing-line value guide, vig and hold guide, bet tracking workflow to keep the examples attached to measurable prices.
Research note board
Use this table to turn the guide into a decision note. The point is to know when the idea is actionable and when it is only context.
| Angle | Input to verify | Example application | Pass when |
|---|---|---|---|
| Market price | Spread, total, moneyline, prop price, or futures hold | Chiefs and Bills compared through PPR | The price has moved past the number that created the edge |
| Football or sport context | Role, pace, weather, injury status, opponent style | Patrick Mahomes role news mapped to the relevant market | The original input changes or remains unconfirmed |
| Review loop | Entry, close, result, and reason code | spreads logged with a clear thesis | You cannot explain whether the process beat the market |
Line movement vs public ticket %
Closing line movement (in points) plotted against the share of public tickets on the favored side. Reverse line moves — where the line moves opposite to public ticket flow — are the canonical sharp-action signal.
Model calibration: predicted vs observed
Predicted win probability bucket vs the empirical win rate inside that bucket on the test set. Points on the y=x reference line are perfectly calibrated; points below mean the model is overconfident in that bucket.
