Here are our NFL preseason numbers, all of them, before you read another word.
- 2023: 9-10, down 1.4 units
- 2024: 10-7, up 1.2 units
- 2025: 9-7, up 0.3 units
- Three years: 28-24, up 0.1 units
Fifty-two picks. Three preseasons. One tenth of a unit. That is not a brag. That is a rounding error, and it is exactly the point.
The formula doesn’t work in August
Our model is built to project how games actually get played. It works in the NFL regular season. It works in college football. It works in the other sports we cover. It does not work in the preseason, and we are not going to pretend otherwise to sell you something.
The reason is simple: preseason coaches are not trying to win the game. They are trying to get through it.
We are not going to open up the recipe here, but the biggest inputs we lose are the ones tied to play calling and substitution patterns, and those two things drive our projected scoring more than anything else. In the regular season, the end of a game is the most predictable part of it. Teams behave in ways you can model. A team down six with two minutes left throws. A team up ten runs the ball and makes you use timeouts.
In August, none of that holds.
A coach up four with three minutes left might kill the clock early because he has decided nobody else is getting hurt tonight. That same coach on a different night might dial up a two minute drill he has no competitive reason to run, because he wants one more look at a third string quarterback and the guy competing to be his fifth receiver. Both decisions are completely rational. Neither one is predictable, and they land on opposite ends of the scoreboard.
Multiply that by 32 coaching staffs with 32 different agendas, and our projected scores stop meaning anything.
The pattern that gets people every year
If you have bet preseason football, you have probably lived this one.
Early in the game, the veteran offenses look sharp, and the defenses look lost, because offenses are simply further along than defenses this time of year. The scoring comes easy. You are cruising to the over midway through the second quarter and feeling pretty good about your read.
Then the brakes come on. The backups come in, the play calling goes vanilla, both staffs decide they have seen enough, and the second half is a slog. Your over dies somewhere in the third quarter.
That is not bad luck. That is the structure of a preseason game, and it happens often enough that you should account for it before you bet, not after.
Week 3 is not safer than Week 1
There is a widely held belief that the preseason gets more predictable as it goes, that the Week 3 dress rehearsal with starters playing a series or two is the one you can actually handicap.
We have not found that to be true. Our confidence does not improve week to week. The variance is not really about who is on the field. It is about what each coach is trying to accomplish that night, and that stays unpredictable straight through. A staff with an unsettled roster spot behaves nothing like a staff that has already made up its mind, and you usually cannot tell which is which from the outside.
What we do about it
Three things.
We put preseason in its own bucket. Preseason results do not touch our regular season record. Not because we are hiding them, they are right at the top of this page, but because mixing them in would misrepresent both. Our regular season record should reflect the formula working. Our preseason record should reflect a market we have told you we cannot beat.
We size small. Preseason plays get conservative units, every time. That is why 28-24 across three years produced a change of one tenth of a unit instead of a real swing in either direction. Less confidence, less money. That is the whole system.
We are selective. Roughly 17 picks per preseason out of a schedule of about 65 games. We are not firing at everything just to have something posted.
So why post them at all?
Because members ask for them, and because we think there is an honest version of what preseason betting actually is.
It is entertainment. That is not a knock. Football is back on television in August after seven months without it, and the back half of most of these games is genuinely rough to watch on a quality of football level. Having a small play on the game gives you a reason to care about a fourth quarter between two rosters that will be cut in half by September. It keeps you watching. That has value.
What it is not is a way to make money. If August is where you are trying to get even, that is worth sitting with for a minute.
Bet small, enjoy the fact that there is football on again, and we will see you in Week 1 when the formula goes back to work.