This Sunday (Feb. 3), the New England Patriots and Los Angeles Rams meet in the Super Bowl, and Americans will gamble an estimated $6 billion on the outcome. There are all sorts of bets to be made this weekend: how many points will be scored, how long the anthem will last, and even what color Gatorade players will dump on the winning coach.
But the most straightforward wager—which team will be victorious—is the most popular. There is a simple way to improve the odds of winning that bet: Go for the underdog.
To attract a (roughly) equal amount of money bet on each team, bookmakers use a tool—the “spread”—to artificially make the odds, well, even. The spread is kind of like a margin of error. It means that if the underdog loses the game, but not by much, gamblers can still win by betting on them. Conversely, the spread works as a handicap for the favorite, who must win by at least a certain amount of points for their backers to succeed.
If two teams are closely matched, then the spread will be small, but if one team is clearly superior, then the spread will be large.
In advance of this year’s Super Bowl, we examined data for more than 9,000 NFL games—in the regular season and playoffs—to test the bookies. It turns out that underdogs have beaten the spread in 52% of games since 1980.
For this year’s Super Bowl, the Patriots are 2.5-point favorites. This means the Pats must win by 2.5 points or more for their backers to win their bets. On the other hand, if the Rams win or lose by less than 2.5 points, their backers get a payday. Of course, there are no half-points scored in football, but for gambling purposes, this half point, the “hook,” guarantees that one side of the bet will win and and one side will lose. (In the event of a “push”—when the spread is matched exactly—gamblers get their money back.)
Our analysis suggests that favorites perform worse when the spread is large. A team favored by 2.5, like the Patriots this year, is only slightly less likely to cover than the underdog; when teams are favored by more than 10 points, they beat the spread less than 40% of the time.
This could be true for a few reasons. First, winning by a lot is harder than winning by a little. Also, in a blowout game, the team with a big lead might not play with the same intensity to close out the game. It’s possible that extreme underdogs play with a chip on their shoulders, putting in an extra effort; instead of getting walloped, they try to make the score “respectable.”
Favorites have beaten the spread within a range of about 40% to 50% of the time (excluding pushes) over the last four decades. During the 2018-19 NFL season, when there were no pushes, favorites beat the spread 44.2% of the time—right in line with history.
In the 52 Super Bowls to date, favorites have fared a little better. They beat the spread 25 times, exactly as many times as underdogs have prevailed (two games ended in a push). Perhaps all the betting on the Super Bowl lets bookies fine-tune the spread, but more likely it’s all just luck. In betting terms, if you want to play the percentages then the underdog is truly the favorite.