On Sunday night, a team that does not play in St. Louis defeated another team that does not play in St. Louis in the championship game of a sport that is not baseball. But I promise we’re going to get there.

In an objectively thrilling, tightly-contested game, the Kansas City Chiefs beat the San Francisco 49ers in only the second overtime game in the fifty-eight year history of the Super Bowl. In what became the most-watched program by total viewership in American television history, and the most watched television event in the United States since the Apollo 11 Moon Landing, the Chiefs won their third championship in five years on a walk-off touchdown from eventual game MVP Patrick Mahomes to Mecole Hardman. Also, Travis Kelce and Taylor Swift continued to exist, a thing I say because even though I’m not in this for the clicks, I can’t pretend SEO isn’t real. Anyway, on to the actual point.

I suspect that most people reading this watched the Super Bowl, or have at least heard about it, but for those who have not, I will recap the opening moments of overtime, as 49ers head coach Kyle Shanahan made a critical and controversial decision before gameplay began. The structure of NFL overtime, particularly with regards to do-or-die postseason games, has evolved over the years. For decades, overtime was completed once a team scored any amount of points, making the decision for what to do upon winning the opening coin toss, Marty Mornhinweg aside, extremely simple–you take the ball and try to win the game before the other team sniffs possession. First in the playoffs and then eventually in the regular season, the NFL adopted a rule that if the team that got the ball first didn’t score a touchdown, the other team would still get a possession–this balanced the scales a little bit, but ultimately the decision remained a no-brainer, including in the previous overtime Super Bowl, during which the New England Patriots received the kickoff and proceeded to drive down the field and score a touchdown before the Atlanta Falcons got the ball.

The new NFL postseason overtime rules have balanced the decision dramatically. The rule now, one semi-ironically crafted as a reaction to a Kansas City Chiefs playoff victory in which Buffalo Bills quarterback Josh Allen was not presented with an opportunity to respond to a Patrick Mahomes-led touchdown drive in what had been a thrilling offensive shootout of a game, indicates that both teams are guaranteed a possession in overtime. If the two teams are tied after having one possession each, the game becomes a sudden-death battle in which the next team to score points wins. So when the San Francisco 49ers won the coin toss on Sunday, this was no longer a de-facto coin toss to determine the game’s outcome, as it had felt in the past–it was a chance to give your team a smaller advantage, but one you should gladly prefer if you’re trying to win any football game, much less the Literally The Super Bowl. The San Francisco 49ers opted to take possession of the ball first.

Because of the unique scoring system of American football, there is a distinct second-mover advantage, and the 49ers, by opting to receive, were forfeiting that advantage to the Chiefs. In college football, where teams alternate possessions starting at their opponent’s 25-yard line, teams that win the coin toss typically choose to play defense first because this allows the team to know what they need to score in order to win or tie the game–a team might be hesitant to kick a field goal with the first possession, knowing that a touchdown by the other team cements them as losers, but it becomes a more exciting option for the team going second if they know already that the first team didn’t score at all. To a lesser extent, this also informs why NFL teams have become increasingly likely to defer receiving the football upon winning the game’s opening coin toss in order to start the second half with the ball.

But unlike college teams that receive the ball first in overtime, there are also distinct advantages that the team receives in professional postseason overtime. The most obvious is that, if the two teams were to remain on equal footing through each’s first possession, the San Francisco 49ers would be in the undeniably superior position of getting the next with a chance to put the game away. But nearly as advantageous is that the looming threat of surrendering the ball to San Francisco necessarily informed Kansas City’s strategy. Once San Francisco had tallied a field goal, the Chiefs were less incentivized to respond with a field goal of their own–in the normal course of play, kicking a field goal on, say, 4th and 9 at the opponent’s 28-yard line is the only sensible choice, but knowing that the best case scenario (making the kick) still leads to you being a prohibitive underdog leads to sub-optimal decisions being made in the name of game theory. Had San Francisco scored a touchdown and followed it with an extra point, the Chiefs would almost certainly have attempted a two-point conversion with a touchdown, since the odds of success are higher than the odds of winning a sudden death game in which they do not start with the ball, but the odds of the two-point conversion are still under 50%, and for their matter, the Chiefs had only attempted (and failed to convert) one two-point conversion all season long.

The San Francisco 49ers made the right decision.

To even say that San Francisco’s decision backfired is a bit of a misnomer. There was very little on Kansas City’s winning touchdown drive to suggest that the same events (though with an extra point attached at the end) would not have transpired in a different order for them. Although this gives the 49ers the knowledge that a field goal is not going to cut it, based on what San Francisco did on their field goal drive, you are requiring Brock Purdy to convert on 4th and 4 and then requiring Brody Purdy and/or Christian McCaffrey to convert a two-point conversion, neither of which are extraordinarily unlikely but in conjunction are not the side of probability on which a team would hope to reside.

The ultimate flaw in Kyle Shanahan’s plan was not that it lacked logic–had this been the problem, there would have been widespread disagreement with the decision in the moment that it happened–but that his team lost the game. And although there are legitimate criticisms of Shanahan’s game in total–notably, a long stretch in which the best running back in the NFL wasn’t given the ball and their ultimate decision to take the overtime field goal when, by going for it, worst-case scenario is giving Kansas City difficult field position–the result was the ultimate problem. *clears throat* And this is where we take a look at baseball.

There have been dramatic tactical changes in how baseball is approached over the last, conservatively, quarter-century. The statistical revolution hit baseball more quickly than it did in other major team sports, and although baseball poets would like to claim that it reflects the intellectualism inherent within the sport, the actual reason is far more institutional–baseball is a series of individual one-on-one events to a much greater degree than other sports, so it is easier to isolate individual performance and moments. But the growing pains associated with this revolution are an ongoing process, not a temporary state which will immediately evaporate. And now it has come to the sporting event by far that is most watched by casual sports fans, in a sport where new metrics have been every bit as resisted, if not more, than in baseball but in a post-Moneyball world.

Sports are frequently compared to chess matches, but unlike chess, a game which has more or less been solved by computers, the numbers of extenuating factors in sports make a perfect decision impossible. A decision can be wholly defensible by probability and deductive reasoning but still fail. And even if the process makes sense, the results can nullify the goodwill towards the sound process.

There is an oversimplification in baseball analytics that suggests that process is what matters and that results should be ignored. I agree with the basic principles that lead to that conclusion, though I do think there is harm in not being at least marginally informed by results, because it works under the assumption that what has been deduced is now gospel. Say the St. Louis Cardinals decided to pinch-hit for Paul Goldschmidt with Iván Herrera and Herrera proceeds to hit a home run. Does this mean that the Cardinals made the right decision? Absolutely not. But if the Cardinals did this ten times and Herrera hit ten home runs, that is probably if nothing else an indicator that Iván Herrera is a better home run hitter than we had previously assumed.

But in the case of the Super Bowl overtime coin toss, the sample size is one–there has been one overtime in which the team chose to receive and that team lost. I suspect, though obviously I cannot prove, that the Kyle Shanahan decision will be the default decision among NFL coaches, but it won’t even be possible for us to get the sample size to two for eleven months. Shanahan’s repeated line, that he chose the ball not because he wanted the ball first but because he wanted the ball third, has become mocked as galaxy-brained because of the inevitability of a Patrick Mahomes touchdown drive (one which would have been his first of the game that went for more than sixteen yards), which doesn’t stand up to the actual probability of a Chiefs touchdown but does stand up to, again, a sample size of one.

In baseball, the sacrifice bunt is nearing extinction because those who studied the run probability charts deduced that the value of an extra base for a runner is not worth the extra out recorded by the defense. It had been largely relegated to being deployed by pitchers in recent years because pitchers had a much higher out probability than position players, but the universal designated hitter accelerated this. But it took literally thousands of sacrifice bunts for this to be considered a bad strategy (aside from using it by surprise, and if the bunter is sincerely attempting to get on base, it is no longer an attempted sacrifice by the literal definition of sacrifice). In football, barring a radical transformation of the rules, you will not live to see a thousand of these new-fangled overtimes–the NFL has had fewer than 700 overtime games in the fifty years since overtime came into place, and even then it would require the NFL to adopt this overtime format for the regular season. Aside from the number of discrete events, the arguable biggest advantage that baseball statheads have is the sheer volume of opportunities to evaluate.

The St. Louis Cardinals were among the first organizations to implement a modern organizational structure, but it could be argued that this progress has slowed in recent years, not because the organization is regressing but because other organizations have accelerated. The Cardinals have moved past some of the most antiquated managerial tactics–they are long past the era of simply putting the fastest guy in the lineup as the leadoff hitter and putting a good bunter second–but this is a team that doesn’t, say, use the bullpen opener. Part of this is likely determined by the age of the team’s players; it was going to be an uphill battle to convince Adam Wainwright, a seemingly smart guy but also a long-tenured creature of habit, of its virtues, and it is unlikely that this year’s starting rotation, which could easily include a 33 year-old Steven Matz as its youngest member, will be much easier to convince.

But this buy-in from the players won’t be the only hurdle for the organization to clear, though it will be the most meaningful one. It will be a challenge to convince fans of the virtues of the bullpen opener when it inevitably makes its debut in earnest in St. Louis not because St. Louis fans are resistant to advanced concepts but because it is inevitably going to fail sometimes. (Hopefully) not most times, but enough that it will be noticed. There is a reason that fans, and this extends to all teams, complain when a starting pitcher is pulled from the game and replaced with a relief pitcher who proceeds to give up a run–we cannot know what the starter would have allowed so we work under the assumption that he will give up runs at his previous rate, which given fatigue will almost never be the case. And the instinct is to get mad.

I am as guilty of falling prey to confirmation biases as anybody. On the increasingly rare occasions when the Cardinals deploy a sacrifice bunt and it doesn’t create any runs, I rail against the out they gave the other team, when statistically it was always unlikely that the bunter was going to avoid an out anyway. This doesn’t mean I was wrong, but it does mean that I am responding to the situation not with reason, even if I have it on my side, but with emotion. And when the San Francisco 49ers received the ball and lost, this was an easy scapegoat, one driven by emotion. It is a phenomenon in sports with which all sports fans should be extremely familiar.