The Sunday before I wrote this column (it will likely be more than a week in the past by the time you read this), the New England Patriots played the Indianapolis Colts in a highly-anticipated night game. It was highly anticipated for a number of reasons: the Patriots were undefeated and had looked extremely dominant for the first few weeks of the season; the Colts had underachieved significantly in their first few games but had recently strung together a few victories and looked to be turning it around. Historically, the Patriots-Colts rivalry has resulted in consistently exciting, if not always competitive, games. Sounds like fun, right? Maybe not appointment viewing, but certainly a decent way to spend Sunday night, especially if you—like me—exhibit certain unfortunate and addictive tendencies re: watching the Patriots.

Of course, that’s not how the game was advertised at all. That’s because of a certain football deflation scandal that I swear I keep trying to stop writing about: it was Ryan Grigson, General Manager of the Colts who, last January, alerted the NFL to potential equipment tampering by the Patriots in the AFC Championship game played between the two teams. Therefore, last Sunday’s game wasn’t just a potentially fun game between conference rivals. It was the Revenge Game for the Patriots—when they, the defending champions, would exact payment in blood from the team that had gotten them fined, humiliated, and had nearly gotten their star player and face of their franchise suspended. The narrative went that the Patriots were angry, and that they were out to make the rest of the league pay. That’s why they had stomped every team they’d played so far.

Well, the Patriots won, but they didn’t exactly stomp the Colts. Indianapolis played well for much of the game, and actually led at halftime. They looked well-prepared to play a Patriots team that, on paper, is much better at almost every position. Apart from a few embarrassing moments including a horribly-designed fake punt monstrosity, it was the kind of ugly-but-competitive football one expects to see after watching enough of the sport. It also went completely against the expected narrative, where, considering the vengeance New England seemed to be wreaking against the rest of the league, the Patriots were supposed to win in a highly-entertaining rout. That didn’t happen, though; the Patriots won, but the narrative lost.

This subversion of narrative happens a lot in football, though. We craft stories to explain why players who are supposed to age out of the league haven’t—“He’s angry that everyone doubted him and is playing harder to prove it!”—or to explain why an underdog can upset a heavily favored team—“They just wanted the win more!” Finely crafted plot arcs like that of the Ballghazi Revenge Game—which, if you wanted to take the time to do so, could be nicely slotted into the classic framework of the Hero’s Journey—end in underwhelming fashion. So why do we do this? Why do fans and media so readily create and accept narratives for football players, teams, and seasons that, historically speaking, don’t often pan out?

Well, we do it in general because people like stories, particularly neatly structured ones. That’s why there are only seven plots. We do it in football, in particular, because the structure of the sport inflates the importance of randomness in game outcomes.

According to ESPN.com, football teams only play 16 games in the regular season, and the average team will run and defend against about 150 plays per game (offense, defense, and special teams), with a little room on either side of that figure depending on the pace of play their team favors, for a total of 2400 plays per team per season. Compare this with, say, basketball, where a team plays 82 games per regular season, with between 90 and 100 possessions per game, for a total of about 8000 plays per team per season. At the individual level, the average starter on a football team takes part in about 1000 plays per season, give or take a few dozen snaps depending on their position, role on the team, and the pace their team favors. With the average play taking about six seconds (although estimates vary from as low as four to as high as eight seconds per play, I’m taking the middle estimate because I don’t want to include plays that are started and then immediately stopped due to penalties or timeouts being taken), that translates to about 134 minutes of total game action for an average starter per season. NBA starters, meanwhile, play about 35 minutes per game, and can log upwards of 2000 minutes of game time per season.

This isn’t to impugn the toughness or hard work of NFL players. Rather, it’s to emphasize just how few plays happen in an NFL season relative to other sports. A whole NFL season can be a very small sample size, statistically speaking. This makes it hard to make predictions with a high success rate, because there: a) isn’t enough data for a given season in the first place, and b) the data you do have could be highly influenced by randomness. In football, it’s really, really easy to misconstrue randomness as something other than chance. To cover this up, we craft narratives. A player isn’t exceeding our expectations because he got lucky or because we were influenced by recency bias; he’s doing so because he wants to stick it to all the haters out there. A team didn’t unexpectedly collapse because they had a lot of bad breaks in a short period of time; they did so because they quit on their coach, or because the locker room was toxic.

Now, having said all that, there’s a caveat: I don’t mean to say that narrative doesn’t have a place in football, or that we’re all trapped in a meaningless, inexpressibly large and chaotic universe that doesn’t care if we succeed or fail. Narratives can be embraced by players and coaches, too, and if a player or team plays with confidence and competence because they buy into a narrative, that obviously influences outcomes on the field. More importantly, even if you take them with a few grains of salt, narratives are fun. It’s fun to see your irrational biases validated. I, as a person irrationally attached to the Patriots, would have a lot of fun watching them become a machine powered by hate, and I definitely want to see them disembowel anyone who had a hand in creating a world where I had to write “Ballghazi” something like 30 times in my most recent column. Even knowing that things could fall apart at any minute, it’s fun to buy into something. And football, in spite of everything, should be fun.