Is there a Super Bowl Loser’s Curse?
That question will confront the Kansas City Chiefs throughout the 2021 season after their defeat last February at the hands of the Tampa Bay Buccaneers. Kansas City is currently 1-1, with a close win over the Cleveland Browns and a close loss against the Baltimore Ravens — both good opponents — and an 81 percent chance of making the playoffs, according to our model. But the Chiefs are also facing a recent trend of regression for teams that came up short on Super Sunday. Both the Los Angeles Rams and San Francisco 49ers failed to make the postseason the year after losing the big game; K.C. is trying to avoid extending that streak to three years in a row.
The bad news for Kansas City: Super Bowl losers do tend to backslide the following season, historically speaking. During their runs to the doorstep of a title, those teams put up a .778 winning percentage (or 13.2 wins per 17 games)1 on average in the regular season, with a +8.9 points-per-game margin. The next year, those numbers dropped to .633 (10.8 wins per 17) with a +5.0 average margin. Relative to league average in each season, Super Bowl losers’ offenses score 2.7 fewer points per game on average year-over-year — thanks in large part to less-effective quarterback play — and their defenses allow 1.1 more points per game.
|Win Pct||PPG vs. Avg||Elo Ratings|
Some of the decline in winning percentage might have to do with lucky runs to the big game — and that luck tending to even out over the following season. During their Super Bowl bids, losing teams won about 1.7 more times per 17 games than we would expect from their point differential (and that’s without taking the playoffs into account). The next year, that difference was down under a half-win, with the teams’ records more resembling their expected wins and losses from the year prior. This factor might be especially pertinent for the Chiefs, who were the luckiest team in the NFL last season in terms of close victories. Kansas City won 4.1 more games than we’d expect from its underlying point differential a year ago, and an 8-1 record in one-score games already gave K.C. prominent placement on lists of potential regression candidates going into Week 1.
Whatever caused their decline, only 14.8 percent of championship losers in history have made it back to the Super Bowl the next season, and just 24.1 percent made it even as far as the conference title game.
What makes Patrick Mahomes so great? | FiveThirtyEight
But sustained excellence is hard for everyone in the NFL, not just Super Bowl losers. (Unless you’re Tom Brady.) If we look at Super Bowl winners rather than losers, their average winning percentage also dropped in the following season (from .789 to .674), in addition to their per-game scoring margin (from +9.5 to +6.4). While those teams were still much more likely to return to the Super Bowl (25.9 percent) or the conference-championship round (38.9 percent) than their losing counterparts, that may have more to do with Super Bowl winners simply tending to be stronger teams to begin with. The eventual winner had the higher preseason Elo rating than the loser in 59.3 percent of all Super Bowl matchups.
In fact, Elo ratings can help us measure the magnitude of any supposed Super Bowl Loser’s Curse, since they account for ordinary regression to the mean by reverting each team toward league average between seasons.2 The year after losing the Super Bowl, teams have historically gone into the following campaign with an average preseason Elo of 1614.1. By the end of the regular season, they wound up with an average Elo of 1605.7 — a shortfall of 8.4 rating points compared with what we’d expect from pure regression.
related: Rejoice, Nerds: Math Won The NFL (In Week 2, At Least) Read more. »
That’s a pretty small effect, considering that the average error between any team’s preseason and end-of-regular-season Elo ratings is plus-or-minus 70.4 points. Based on the normal distribution of Elo’s errors, we would expect a team with a given preseason Elo to fall at least 8.4 points below their predicted mark around 46 percent of the time — hardly the kind of unlikely shortfall we’d expect from a full-blown curse.
Now, the 2019 Rams and 2020 49ers each missed their expectations by quite a bit more than the historical average, which has given the Super Bowl Loser’s Curse new relevance. But before that, it was looking like a pretty dead phenomenon. Yes, the Carolina Panthers went 6-10 the year after going 15-1 and losing to Denver in Super Bowl 50, but nine of the 10 previous Super Bowl losers before 2018 would return to the playoffs the following season, and nine of 10 also had a winning record. And while only one of those teams — the 2017 New England Patriots — made it back to the Super Bowl, they happened to win it all (over the Rams).
Before that, the curse seemed more real. Only two of the 10 Super Bowl losers leading up to 2009 returned to the playoffs, and only three posted winning records.3 But in the 10 seasons before that, seven title-game losers returned to the postseason (including the Buffalo Bills, who actually went back to the Super Bowl itself in 1991, 1992 and 1993 — granted, only to lose each time). And before that, every single Super Bowl loser from 1969 to 1986 at least went back to the playoffs.
Like most sports curses, the Super Bowl Loser’s Curse mainly seems to be just another term for regression to the mean: outlier performances being followed by slightly less impressive encores, with a variety of causes. There are legitimate reasons to think the Chiefs may return to a third straight Super Bowl (Patrick Mahomes continues to play like Patrick Mahomes) or that they may not (their defense has ranked among the NFL’s worst in the early going). But whatever happens to Kansas City in 2021, it probably won’t be because of a hangover from what happened at the end of last year’s playoffs.
Check out our latest NFL predictions.