Same bet, different price
A player prop is a bet on one player's numbers rather than on who wins. Will a receiver catch more or fewer than 4.5 passes? Will a running back go over or under 62.5 rushing yards? Every major sportsbook posts these, and most bettors take whatever their own book is offering.
That habit has a cost, and we can measure it. MatchWiz records the closing price every book hung on every NFL player prop before kickoff, going back to the start of the 2023 season. That's 902 games. We took every bet that at least three books quoted on exactly the same terms (same player, same market, same line, same side) and asked a simple question: how much better was the best price than the typical one?
There were 45,790 of those bets. On 89% of them, the best book paid more than the middle of the pack.
How much more, in plain terms
Odds look like −110 or +105, but the cleaner way to compare them is what a winning $100 ticket pays back. The typical prop in our sample sits right around −110, which returns $190.90 on a $100 winner.
The best available price paid about 1.9% more than the median one on the typical bet, and 2.4% more on average. Nearly three bets in ten were at least 3% better at the best book than at the middle one.
Two percent sounds small until you remember how thin the margins in betting are. At −110, you need to win 52.4% of your bets just to break even. At the best price on the same bet, the break-even rate drops to 51.4%. That full point is roughly the difference between a good bettor and a break-even one.
Put numbers on it. A bettor who wins 54% of their props at the typical price earns about 3.1 cents of profit per dollar wagered. The same bettor making the same picks at the best price earns about 5.1 cents, close to two-thirds more profit for no extra skill. A bettor winning 52% of the time loses about 0.7 cents per dollar at the typical price and makes about 1.2 cents at the best one. Same picks. Opposite result.
Books don't even agree on the number
Price is only half of it. Often the books disagree about the line itself.
We looked at 29,130 cases where three or more books posted an over/under line for the same player and market. In 57% of them, at least one book's line was different from the others. When they disagreed, the typical gap was two full units, which on a yardage prop means two yards.
How often that happens depends heavily on the market. Passing yards lines differed across books 95% of the time, receiving yards 85% and rushing yards 81%. Counting markets were far more settled: receptions lines differed 22% of the time and passing touchdowns only 11%.
A two-yard difference on a receiving total isn't cosmetic. If you like the over, 54.5 and 56.5 are different bets, and the half-points in between are where a lot of these props are decided. Our price comparison above only counted bets on identical lines, so it understates what shopping is worth. The better line is a second, separate edge.
Which markets reward shopping most
Among the identical bets, the gap between best and typical price was largest on passing touchdowns (3.2% on average), receptions (2.9%) and rushing attempts (2.7%). It was smallest on passing yards (1.4%), where the books compete hard on price but, as above, disagree most on the line.
That pattern makes sense. Counting markets like receptions settle on whole numbers, so books tend to post the same line and fight over price. Yardage markets have room to move the line itself. Either way, the bettor who checks only one book leaves something on the table.
Anytime touchdown props are their own case. Books only offer a yes on those, never a no, so part of the gap between them is just each book's built-in margin rather than a pure difference of opinion. With that caveat, the best price on an anytime touchdown paid a median 11% more than the typical price across 22,656 comparisons, and more than half were at least 10% better. If there's one prop where checking every book matters most, it's this one.
No single book is always best
The natural question is whether one book is simply the best place to bet. It isn't.
FanDuel held the best price, or tied for it, on 44% of the bets it quoted. DraftKings did on 39%, BetMGM on 34% and Caesars on 18%. Put the other way around, even the book that won most often was beaten more than half the time.
A bettor loyal to one book gives up, on average, about 2.0% per bet against the best price at FanDuel or DraftKings, 2.7% at BetMGM and 3.3% at Caesars. Spread across a season, that is the same kind of gap we described above, paid quietly on every ticket.
This is also why MatchWiz grades every play at the price that was actually posted when we called it, rather than at the best price we could find later. Grading at a cherry-picked price would flatter our record by exactly the amount this piece measures.
What this doesn't say
It doesn't say any sportsbook is a bad one. Every book in the sample held the best price some of the time, and each prices different markets more aggressively than others.
It doesn't say shopping turns a losing bettor into a winning one. Two percent closes a small gap, not a large one. Someone hitting 48% of their props will still lose, just more slowly.
It measures closing prices only, the last number each book hung before kickoff. Prices earlier in the week can differ by more or less, and a line can move between the moment you look and the moment you bet.
It assumes you can bet at several books, which depends on where you live. Some states have only a handful of legal sportsbooks, and the gap shrinks with fewer to compare. It also ignores betting limits, which some books apply to winning customers.
And it's one sport and three seasons plus the start of a fourth: 902 NFL games, 45,790 identical bets, up to eight books per bet. We'll re-run it as the sample grows and publish whatever it says, including if the gap narrows.
Sources
Every number here comes from the graded ledger the boards publish. Research from a statistical model, not betting advice; no outcome is guaranteed. Method: how MatchWiz works.