Does NFL Preseason Betting Have More Value Than the Regular Season?
Open a sportsbook app today and you can bet Saturday’s Cowboys-Seahawks preseason game or a Week 12 regular-season game that does not kick off until late November. Both are priced, on the same screen, at the same book. The standard argument for NFL preseason betting value says the August market is the soft one: thin lines, distracted oddsmakers, a product nobody is really watching. We checked that against our own odds archive, and the pricing half of the argument does not survive contact with the data.
Through the 2026 preseason, DraftKings and FanDuel have charged the same margin on preseason games that they charge on regular-season games, and they have disagreed with each other less than they do on the regular season. Whatever edge August offers, you are not getting it at a discount. You are getting it from knowing who is actually going to play.
What “Value” Actually Means When You Bet the Preseason
Value in a betting market comes from two separate places, and the preseason argument almost always blurs them together. The first is the toll you pay to get in the door: the hold, the vig, the margin baked into the two sides of a price. The second is accuracy: how often the number itself is wrong. Those are independent. A market can be cheap to enter and priced perfectly, and it can be expensive to enter and badly wrong.
The toll is easy to measure, because it is arithmetic. A standard point spread priced -110 on both sides carries a hold of 4.55%. Take the implied probabilities of both sides, add them up, and the amount by which they exceed 100% is the book’s cut. Accuracy is much harder to measure, because it requires knowing what the true number should have been, which nobody does.
So the honest version of the question is two questions. Do books charge you more to bet the preseason? And are preseason numbers more often wrong? We can answer the first one with real data. The second one has a real answer too, and it is not the one the trend articles give you.
We Measured the Vig on Every 2026 Preseason Game. It Is the Same.
DraftKings and FanDuel have priced the 2026 NFL preseason at the same margin they are charging on the regular season. Across every pregame snapshot our odds archive captured in August 2026, preseason spreads held 4.58% and preseason totals held 4.54%, against 4.56% for both on regular-season games from the same two books in the same month.
| Market | Preseason hold | Regular-season hold |
|---|---|---|
| Point spread | 4.58% | 4.56% |
| Total (over/under) | 4.54% | 4.56% |
| Moneyline | 3.96% | 4.01% |
Look at where the spread and total numbers land: 4.54% and 4.58% sit right on top of that 4.55% figure a -110/-110 line produces. The books are charging standard football pricing on exhibition games. There is no August surcharge, and there is no August discount either.
📊
How we measured this
Our odds archive logged 17,049 NFL price rows in August 2026. We used pregame captures only (anything at or after kickoff is a live in-game price and would distort everything), and we restricted the comparison to DraftKings and FanDuel, requiring both to be pricing the same game at the same moment, which holds the book set fixed on both sides of the comparison. Two limits worth stating: the regular-season side is games priced weeks or months before kickoff, not an in-season Week 8 line, and our capture cadence runs roughly every six hours, so a move that happens and reverses inside a single window is invisible to us.
Do Sportsbooks Disagree More on Preseason Lines?
They disagree less. On the same games at the same moment, the two books sat closer together on preseason numbers than on regular-season ones across all three markets we track:
- Point spread: 0.25 points apart in the preseason, against 0.52 points apart on regular-season games.
- Total: 0.33 points apart, against 0.56 on the regular season.
- Moneyline price: about one percentage point of implied probability apart, against roughly 1.4 points on regular-season games.
That matters, because cross-book disagreement is the mechanism that makes line shopping pay. When one book has a game at -3 and another has it at -3.5, somebody is wrong and you get to pick a side. In the 2026 preseason, the two books have mostly been sitting on the same number. Shopping is still worth doing (it always is), but the gaps you are hunting are smaller in August than they are on a regular-season game the books have not sharpened yet.
None of this means the preseason number is right. It means the books are confident enough in it to stand close together. Whether that confidence is earned is the interesting part.
Do Preseason Lines Move More Than Regular-Season Lines?
Before kickoff, they barely moved at all in Week 1. Across all 16 preseason Week 1 games, the median pregame range on the consensus spread was zero points. The average drift was 0.30 points, and the single biggest pregame move we recorded was 1.8 points, on Chargers at Texans.
This is where a lot of preseason folklore comes from, so it is worth being precise about what we are looking at. If you glance at a preseason game’s line history and see it swing from -3 to +12, you are almost certainly looking at live in-game odds from a game that got out of hand in the third quarter, not at a pregame market being hammered by professionals. When we filtered our archive down to captures taken strictly before kickoff, those dramatic swings vanished. Ten of the 16 Week 1 games have not moved off their opening number at all.
Our six-hour capture cadence is a genuine limitation here, and we would rather flag it than pretend otherwise. A sharp move that lands at 2pm and gets bet back by 6pm would not show up in our snapshots. But a market being repeatedly moved multiple points by informed money should still leave a mark across a two-day window, and in Week 1 it did not.
So Where Does the Preseason Edge Actually Live?
In playing time. A preseason result turns mostly on how long each coach intends to use his starters and his likely Week 1 rotation, and that intent gets disclosed unevenly in the days before kickoff. The talent gap between two rosters is already in the number. How much of that talent actually takes the field is the part the number struggles with.
Preseason Week 1 was a clean illustration of how wide that split gets. Per CBS Sports’ rundown of every Week 1 game, coaches went in opposite directions on the same weekend:
- Playing their starters: Cincinnati (Joe Burrow and Ja’Marr Chase), Green Bay (Jordan Love), Carolina (Bryce Young), and Cleveland, which put its first-team offense on the field for the opening series.
- Sitting them: Detroit, where Jared Goff did not play and Dan Campbell held back most of the starting group; Jacksonville, where Liam Coen ruled his starters out entirely; Kansas City, where Andy Reid called Patrick Mahomes unlikely; and Dallas, where Brian Schottenheimer said plainly that the Cowboys would not be playing many starters at all.
- Sitting them all preseason: the Rams, who as a matter of policy do not play their starters in exhibition games.
That is the real asymmetry, and it is a research problem rather than a pricing problem. The coach who announces on Wednesday that his quarterback is playing a series and a half has told the market something. The coach who says “we’ll see” has not, and the number has to guess. If you are going to bet these games, that gap is where your time goes: beat reporters, Wednesday and Thursday press conferences, and depth-chart reporting from people who cover the team every day.
💡
The second- and third-string question
Preseason games are decided in the second half by players fighting for the last few roster spots, so depth matters more than star power. A team with a genuinely competitive backup quarterback battle often plays a sharper fourth quarter than a team whose third-stringer is a camp arm. That is the sort of detail a market built on team reputation prices poorly, and it is checkable before kickoff.
Why Preseason Betting Systems Deserve a Hard Look
The NFL plays 49 preseason games a year, which is not enough football to support most of the trends built on it. Three weeks of 16 games plus the Hall of Fame Game is the entire annual sample, and it has been that small only since 2021, when owners approved the 17-game regular season and cut the preseason from four games to three to stay inside the CBA’s 20-game cap.
Run the arithmetic on what that does to a “trend.” A system that only fires on road underdogs of 2.5 points or less might catch six or eight games a season. Ten years of that is well under a hundred bets, and the rule changed partway through, and the coaches turned over, and half the sample came from a four-game format that no longer exists. A 60% hit rate on 80 games is roughly what you would expect a coin to produce a few times if you tested enough different coin-flipping rules, which is precisely what a trend database does.
⚠️
The tell to watch for
If a preseason system comes with three or four qualifying conditions stacked on top of each other (road team, off a win, favored by under a field goal, in the second week), the conditions were almost certainly chosen after somebody looked at the results. That is curve-fitting, and it does not carry forward. A rule that needs that much shaping to look profitable is describing the past, not predicting the next game.
Coach-by-coach preseason records have the same problem in a worse form. A head coach with a strong August record accumulated it across maybe 20 or 30 games, often at a different franchise with different personnel and a different depth chart. We publish our own results monthly in the closing line reports precisely because a record needs a real sample and a public method before it means anything, and preseason trend tables almost never have either.
The Limits Problem: A Real Edge You Cannot Scale
Even a genuine preseason edge runs into a ceiling, because books cap what you can put on these games. This is the oldest and most reliable observation about the preseason market, and it cuts directly against the “August is where the money is” pitch.
ESPN reported the shape of it back in 2018, quoting Westgate SuperBook oddsmaker Ed Salmons, who put preseason limits around $2,000 against five-figure regular-season limits, and estimated that roughly 80% of preseason handle came from professionals versus about 20% in the regular season. Those are 2018 figures from one Nevada book and we would not present them as today’s numbers at a national app. The structure has held up, though, and the logic is obvious: a book facing a market where most of the money is informed and the information is cheap to acquire protects itself by shrinking the bet size rather than widening the price.
That reframes the whole question. A market where the sharpest bettors make up most of the handle is not an easy market. It is a market the recreational money has left, which is the opposite of the condition that usually creates soft lines. And the practical consequence is that even if your Wednesday depth-chart research is genuinely good, you are grinding it out in small units against a field that is mostly other people doing the same work.
How to Approach the Rest of the 2026 Preseason
Treat it as a small-stakes research market, not a bankroll opportunity. Two weeks of games remain before the preseason closes on Saturday, August 29, 2026, and the regular season opens Wednesday, September 9, 2026 with the Patriots at the Seahawks in a rematch of Super Bowl LX. That is a short runway, and the sizing should reflect it.
- Do the playing-time work first, or skip the game. If you have not read this week’s beat reporting on both teams’ rotation plans, you are betting a number built on rosters that will not be on the field.
- Weight totals over sides. A total absorbs both teams’ playing-time decisions at once, and a game where both coaches are sitting starters has a much more predictable scoring profile than a spread does.
- Fade the reputation, not the record. The market’s price on a preseason game still carries some residue of how good the team is in September. Depth and roster-bubble competition decide these games.
- Expect small limits and do not fight them. If a book caps you at a few hundred dollars, that is the market telling you what it thinks of your side.
- Shop anyway, but know the gaps are thin. Our data has DraftKings and FanDuel a quarter-point apart on preseason spreads, so the value in shopping August games is mostly in the price, not the number.
The honest answer to the headline question is a qualified no. NFL preseason betting does not carry more value than the regular season in any sense you can see on a price screen: the vig is the same, the books agree with each other more, and the lines do not swing the way the folklore says.
What August offers is a narrower, more specific opportunity, available to a bettor willing to spend Wednesday reading depth-chart reporting and willing to accept that the payoff comes in small units. That is a real thing. It is just not a discount, and anyone selling it as one is selling you a trend table. If you want our current work on games we do rate, it is on the picks page.
Play Safe: Gambling should be fun, not stressful. Set limits, stick to your budget, and never chase losses. If you or someone you know has a gambling problem, call 1-800-MY-RESET or visit ncpgambling.org. For more resources, see our Responsible Gambling page.
Frequently Asked Questions
The questions we get most often about betting NFL exhibition games, answered with what our odds data actually shows.
Is NFL preseason betting actually worth it, or is it a trap?
It is worth it only if you are willing to do playing-time research before every bet, and only at small stakes. Our August 2026 odds archive shows DraftKings and FanDuel charging the same margin on preseason games as on regular-season games (4.58% on preseason spreads against 4.56% on regular-season spreads), so there is no pricing discount to collect. The opportunity is in knowing which starters will play and for how long, which takes real work and pays off in small units because books cap preseason bet sizes.
Do sportsbooks charge more vig on preseason games than regular-season games?
No. We measured every pregame price our odds archive captured in August 2026 and preseason spreads held 4.58%, preseason totals held 4.54%, and preseason moneylines held 3.96%. The comparable regular-season figures from the same two books were 4.56%, 4.56% and 4.01%. A standard -110/-110 line carries a 4.55% hold, so preseason football is being priced at ordinary football juice.
Why are the betting limits so much lower in the preseason?
Because most of the money bet on preseason games is informed money, and books manage that risk by shrinking bet size instead of widening prices. ESPN reported in 2018 that Westgate SuperBook oddsmaker Ed Salmons put preseason limits near $2,000 against five-figure regular-season limits, and estimated roughly 80% of preseason handle came from professional bettors versus about 20% during the regular season. A market dominated by sharp money is a hard market, not an easy one.
Should I bet preseason totals instead of sides?
Totals are usually the better place to start, because a total absorbs both teams’ playing-time decisions at once. If both coaches are resting starters, the scoring profile becomes fairly predictable in a way the spread does not. Sides require you to be right about the relative depth of two specific backup units, which is a harder call to make from the outside.
How do I find out whether a team’s starters are actually going to play?
Head coaches usually say so at their Wednesday or Thursday press conference, and local beat reporters publish it immediately. In Preseason Week 1 of 2026 that produced a clean split: Cincinnati, Green Bay, Carolina and Cleveland played starters, while Detroit, Jacksonville, Kansas City and Dallas held theirs back, and the Rams do not play starters in any preseason game as a matter of policy. Following the beat writers who cover both teams is the single highest-value habit in preseason betting.
Do those preseason betting trends and coach ATS records actually mean anything?
Treat them with real skepticism. The NFL plays only 49 preseason games a season, and it has been that few only since 2021, when the league cut the preseason from four games to three. Any trend narrow enough to sound convincing (road dogs of under a field goal, teams off a blowout win) is running on a few dozen games spread across rule changes and coaching turnover. The more qualifying conditions a system stacks on, the more likely those conditions were picked after somebody saw the results.
