Betting the NFL Preseason

Preseason NFL football gets a surprising amount of betting action. In fact, many sports books report there's more money wagered on the Hall of Fame Game (the first preseason game) than on the rest of that day's events combined, which includes a summer Sunday of baseball, tennis, golf, and summer league basketball.

NFL teams play four preseason games during an excruciating mini-season in August and early September. Teams typically will play twice at home and twice on the road, but there are occasional promotional games in non-NFL cities.

There is serious talk among the powers-that-be of finding ways to reduce the number of preseason games without losing the associated gate and media revenue with the two home games each owner gets per year. One proposal is to trade preseason games for a longer regular season. But, the NFL Players Association appears to be solidly against adding two regular season games. It will all come down to the renegotiation of the Collective Bargaining Agreement between players and team owners.

What makes preseason NFL football so underwhelming as a fan is the uncanny valley: It resembles a real NFL game in every way, from the announcing voices to the stadiums to the uniforms, but something is just disturbingly out of whack. The new normal is to keep veteran players quarantined. Stars often play no more than a series in the first few games and not at all in the last preseason game.

Thankfully, we have gambling to keep us cool in the dog days of summer. Preseason football doesn’t have to be a choice between the lesser of boring and who cares. In a meaningless game, populated mostly by spare players running a playbook straight out of last year’s Pro Bowl (erp), at least you have the ability to make it interesting through a wager.

Preseason lines and limits

The fact that sharps make bets on the preseason should tell you that it’s worth pursuing. In fact, sharp action often outweighs the public on individual games, and that means you’ll see bigger odds moves, and when it happens, you can derive more meaning from it because it’s less likely to be fueled by the tourists and occasional squares.

If you are a high-dollar bettor, you’ll notice that preseason betting limits are much lower than regular season limits. Some bookmakers limit bettors to 20 percent or even 10 percent of a maximum bet they might take in the regular season. Bookmakers are constantly trying to balance opportunities for increasing handle against the risk of getting beaten, and they have extra risk in high-variance games. Bookmakers want contests between two teams that put winning above nearly everything else. Because preseason games are meaningless, team and coaching motivation can be hard to pin down, which means outcomes are harder to model. Preseason carries an elevated risk of a single piece of information about the game might confer a big gambling advantage (like a coaching staff’s plans for playing their 1st string quarterback), and there’s nothing oddsmakers hate worse than losing the information battle with gamblers.

We know the NFL regular-season side and total markets are efficient: The oddsmakers models are accurate, there are lots of bettors who process and bet on new information, and oddsmakers quickly adjust lines. A good research idea might be to compare the efficiency of the preseason and regular season betting markets. With lower limits and more professional bettors, are there any systematic weaknesses in the preseason market that aren’t in the regular season market?

You can get a sense for how efficient the market is by measuring the average distance between closing lines and winning game margins. This is sometimes called the spread margin. The lower the spread margin, the closer the game margin was to the spread. In recent seasons, the preseason NFL number has actually shown itself to be tighter (9.2) than the spread margin for regular season games (9.5).

So are preseason games as hard to beat as regular season games? The tighter spread margin difference between regular and preseason could be attributed to swings in points scored. (If an oddsmaker sets a game at a pick’em, you’d see a lower spread margin if the game ended 42–0 rather than 44–0, but it would be hard to say that the former score represents a materially better spread on the game). And the golden rule of bookmakers is that they lower their betting limits when they feel vulnerable. That makes me think the preseason is worth looking into if you’re a bettor.

Preseason as predictor

Let’s start with a basic question about NFL team performance: Do teams that win in the preseason go onto win in the regular season? The answer appears to be no. Writer Chad Langager did a study a few years ago where he looked at ten seasons of data and found very little correlation between preseason winning percentage and the subsequent regular season winning percentage.

But of course, we're not trying to predict regular season winning percentages. We're looking for betting opportunities in the preseason.

Betting on head coaches

The surest spread play in the NFL preseason revolves around coaching tendencies. If you played a sport competitively, then you undoubtedly had a coach tell you that it was important to "practice like you play" and that winning had to become a way of life, not just an outcome. On the flip side are the coaches who treat preseason as the fans see it: basically meaningless, and wins are certainly not worth risking player health over. I can see both sides of the argument, but really, I’m just glad there are coaches out there who can be so easily profiled.

Coaches come and go, but there are some who stand out when it comes to the preseason.

Preseason win-obsessed:

• John Harbaugh
• Mike Zimmer

Preseason win-oriented:

• Bill Belichick
• Jon Gruden
• Mike Tomlin
• Pete Carroll

Preseason losers:

• Jason Garrett
• Dan Quinn
• Matt Patricia

Jury’s still out:

• Doug Pederson
• Sean Payton
• Ron Rivera

I separate out Zimmer and Harbaugh because together they have won nearly two-thirds of their preseason games and have spread records to match. The approach is simple: Bet on the coaches who like winning; bet against the ones that have losing tendencies (if they’re still in the league); and on the occasions when one type plays the other, bet a little bit bigger. And in case you’re worried about tracking coaches and their straight-up winning percentage, understand that preseason spreads rarely stray beyond 4 or 5 points, so the straight-up winner is usually the point spread winner as well.

If you’re reading this article in 2020 or beyond, there will inevitably be some new names in the coaching ranks, and some of the guys in the above list will have faded away. More important, some of the coaches I didn’t list will have accumulated enough data to reveal themselves to be a Zimmer-type or a Garrett-type. Before August comes around, do yourself a favor and look up the current coaching crop’s preseason ATS record.

Considering the coaching angle is so widely discussed, it’s surprising that it has not been priced out of the market yet. During the 2019 preseason, some analysts claimed the Zimmer-coached Vikings and Harbaugh-coached Ravens had more expensive point spreads than they normally would have. Maybe so, but they both continued to win.

Other potential factors related to coaching include coaches on the hot seat. It’s been theorized that they’re more likely to try and get wins in the preseason. After all, when you’re fighting for your job, every non-losing result helps. In an admittedly unscientific look at this, I counted 7 “consensus” hot seat head coaches going into the 2019 season. I threw out two of them (Garrett and Zimmer) who had already been profiled. That left 5 hot seat coaches who, drum roll please, went a combined 7–14 in the preseason. Clearly this needs a deeper look.

BET on week 2 losers

Systems are hard to come by for preseason NFL, but this here’s one that’s intuitive, well known, and surprisingly has not been squeezed out of existence:

In the NFL preseason

Both teams are playing their second preseason game AND

One of the two teams lost their first preseason game outright AND

The other team won their first preseason game outright

BET the point spread of the team that lost last week.

Statistically, the correlation between the winning percentage of a team in the preseason and their winning percentage in the regular season is 0.0944. Values between –0.1 and +0.1 suggest no correlation between two variables.

Other preseason betting factors

Here’s a list of basic factors to consider when assessing a preseason game:

• How long will starters play? Most coaches have playing time recipes determined well before the game. Look for games where one team plans to leave starters in longer than the other; consider a 1st half or 1st quarter wager.
• The 2nd string quarterback is often the pivotal player when it comes to preseason results. Beyond the quarterback, look at differences in overall depth of bench talent for the two teams. In the salary cap era, it’s hard for teams to be both good and deep at the same time.
• In preseason games 3 and 4, watch for skill position players on the bubble. If the team is only planning to keep two quarterbacks, find prop bets to take advantage of a third-string quarterback who’s likely to be going for the big splash to save his skin. Interceptions anyone?
• Skip it. Spare yourself from watching it. Do some calisthenics. Play with your kids. Call your mother.