Warning: This post was written by a Mets fan.
When Daniel Murphy let a ball bounce beneath his glove in the eighth inning of Game 4 of the World Series, I threw my Mets hat to the ground. It was ostensibly the pivotal moment in a 5-3 Royals win, the kind of play that reminded me why other sports have fouls but baseball has errors.
We could sit here together and dwell on all the Mets errors: We could wonder how Jeurys Familia, an all-star closer, blew five saves in 48 opportunities during the regular season, but has blown two in two opportunities during the World Series. We could plumb the depths of Yoenis Cespedes’s above-average defensive metrics, and make the case that to measure Cespedes’s true defensive capacity we need a new metric that somehow quantifies a fielder’s nonchalance.1 And we could spend hours trying to understand Terry Collins’s faith in an eighth-inning set-up man who is allowing a .835 OPS to opposing batters in the postseason. (It seems as though I may spend the next several years doing that.)
But instead I want to tell you about my hat. If you read FiveThirtyEight a lot, you know that we’re puritanical about baseball’s playoffs being a crapshoot. They’re a series of games that may or may not be a reflection of a team’s actual quality. Intellectually, I know the same rules of randomness that apply to a baseball also apply to what I wear to watch a game. But the World Series is not a time for intellect.
On July 31, the day Cespedes was traded to the Mets, I bought a Minnesota Twins hat at Target Field in Minneapolis. It was a tourist’s purchase – I was in Minnesota for a couple of ballgames with some friends.
But the hat started to mean something more. That weekend, the Mets swept the Washington Nationals to tie for first place in the NL East. So I kept wearing the hat. And the Mets kept winning. The Mets went 37-22 to close out the season, and won the NL East despite a 23 percent chance of doing so when I bought the hat. (The rational readers among you will note that they also went 37-22 to close out the season after Cespedes joined the team, but, again, this is not a rational story.)
Soon, the Twins hat had replaced my Mets hat. My Mets friends texted me and asked me to wear it when they were feeling nervous about a game. I nearly forgot it on a plane, and felt the Mets season slipping away until I stormed back to retrieve it. At the start of the playoffs, I went on a poorly timed vacation to India, and brought the Twins hat to ensure the Mets advanced.
I returned to the U.S. in time for the World Series, and there was no question I’d wear the Twins hat into the heart of a Mets bar for Game 1. Fourteen innings later, the hat wasn’t enough. The Mets lost 5-4.
So I put on something different. I went to Game 3 in Citi Field and wore a hat that spelled out M-E-T-S. I had worn it to every Mets home game I attended this year.
That Mets hat has its own history, with a winning percentage of about .550 this season, if I recall correctly. Good, but not Twins hat good. Yet the Mets won Game 3 9-3. And so, before Game 4, I faced the same choice any manager does: Do I ride what’s hot, or stick with the steady performer? I looked into the archives of Baseball Prospectus, but couldn’t find any research on whether there’s such a thing as a hot-hand effect in fans’ attire. I was adrift with nothing but my own small sample sizes.
Saturday, I put on the Mets hat. By the end of the night, it had regressed to its mean. It couldn’t stop a Royals team that had a .301 BABIP in the regular season from having a BABIP of .346 in Game 4 (and that doesn’t even count the ball that skittered beneath Murphy’s glove). It couldn’t make the Mets win. It couldn’t get Terry Collins to bring in his best bullpen pitcher for a six-out save with nobody on, rather than a five-out save with two runners on base. It couldn’t get Cespedes to stay closer to first base in the final moments.
A hat with a .550 winning percentage could never do that. But a Twins hat with a .627 winning percentage? We’ll find out during Game 5.