What Microwave Burritos Have in Common With Postseason Success
As the man who inspired Brad Pitt’s most memorable role once said, “My shit doesn’t work in the playoffs.” Assuming Billy Beane wasn’t explaining an October Metamucil purchase to a grocery store cashier who simply asked how his day was going, what Beane likely meant was that the statistics used to construct his major league rosters don’t accrue large enough samples during postseason series to eventually even out in his favor. Over the course of 162 games, a team’s production settles into a reasonable representation of the squad’s true talent. But zoom in on any random seven-game stretch and the team on the field might look like a bunch of dudes in baseball player cosplay.
What applies to team outcomes applies just as well to player outcomes. A player with a perfectly respectable stat line in the regular season might morph into a pumpkin as the calendar shifts to fall, or on the flip side, an unlikely hero may emerge from the ashes of a cruel summer and put the whole team on his back.
With the law of averages in mind, I’d always assumed that the more consistent hitters would be better positioned to perform well in the playoffs. My thinking went like this: The natural variation in these hitters’ performances would never wander too far from their season-long average, making them the safer, more predictable options. Whereas streaky hitters — the ones with high highs, low lows, and steep transitions between the two — would be too reliant on “getting hot at the right time” to be the type of hitter a front office should depend on in the postseason.
Reader, I was incorrect.
As it turns out, the postseason likes its hitters like it likes its microwave burritos: blazing hot one bite, ice cold the next. That’s not to say that the postseason turns its nose up at the microwave’s other offerings, but rather than sticking to foods like mac and cheese that deliver a uniform temperature through the simple act of stirring, October baseball insists on imperiling its taste buds on pizza rolls and Hot Pockets and whatever other dough-encased compilations of meat and cheese appear in the frozen food aisle. You know, the types of foods for which each bite reveals either lava straight from the depths of Hell or a substance masquerading as a discarded flavor from the Dippin’ Dots test kitchen.
To determine the microwaveable entree most likely to satiate the beast that is postseason baseball, I first had to group hitters into categories. To differentiate between the streaky hitters (to whom I may subsequently allude using various burrito nomenclature) and the consistent hitters (which is the group I’m talking about whenever you see a reference to mac and cheese, or MAC for short), I borrowed some methodology from Justin Choi, where he compared the amount of variation across weekly segments of offensive output. The highly variable ones are the streakiest, while those with minimal variation are the consistent ones.
Though I kept that general framework, I did make a couple of tweaks. Instead of using calendar weeks, which provided somewhat arbitrary endpoints that may not fully represent the magnitude of a hot streak or cold spell, I used a rolling, seven-day average (minimum 20 plate appearances), then sorted the samples best to worst and alternated selecting from the top and bottom of the pile to get each player’s best and worst weeks. If a selected seven-day sample had more than a four-day overlap with any of the previously selected samples, it was tossed out.
And because I was tabulating weekly averages myself, I used wOBA rather than wRC+ to simplify the process even though it did mean sacrificing the park factor adjustment. (Don’t worry though, the top of the Team Burrito leaderboard is not littered with Rockies rotating between homestands and road trips.)
Variation across weeks was measured using standard deviation. Hitters with values in the top 25% of standard deviation went in the streaky category and hitters in the bottom 25% landed under the consistent label. From here, I took the hitters from both groups and compared each individual’s regular season wOBA to their postseason wOBA (only including players with at least 20 PA in the postseason), then averaged the difference across the entire group. To zone in on the players upon whom teams are most reliant in the postseason, I filtered the data set to hitters with a .330 wOBA or higher. Because if a team’s plan to win the World Series hinges on a player with a .290 wOBA getting hot, it probably has other lineup issues to address before worrying about the temporal distribution of offensive production. Like maybe work on getting some offensive production period.
Last thing before we get to results, I grouped individual player production two ways when comparing regular season performance to playoff performance. One version compares each player’s regular season wOBA to postseason wOBA within a single season before averaging across the entire group, while the other compares each player’s career numbers (though I did only pull in data from the last 10 years, so some players’ full careers didn’t make the data set). In the single season version, players are classified as Burritos or MACs based on that individual season, while the career version makes that determination based on the player’s entire body of work.
When using single seasons of data, both Burritos and MACs are equally likely to post numbers similar to what they did in the regular season. A player with a .350 wOBA through September is just as likely to swing his way to a .350 wOBA in October regardless of whether that .350 wOBA came together while careening through a season of peaks and valleys or gliding across a field of gentle rolling hills. This tells me one of two things, either a single season isn’t enough to determine a hitter’s archetype, or archetype has no impact on a player’s ability to perform to his own standard in the playoffs.
That’s where grouping by career numbers comes in. As I looked at the single season data points, I noticed several players showing up multiple times in both categories, depending on the season. Ronald Acuña Jr. is classified as streaky in 2018 and ’19, but consistent in ’22 and ’23. In all four seasons he hovers pretty close to the boundary for earning either label. In 2021, he falls in neither camp. Similarly, Kyle Schwarber hangs out at the consistent end of the spectrum in 2017 and ’18, before setting up permanent residence on the streaky side of things from ’19 onward. When looking at Schwarber’s seven-game rolling wOBA throughout his career (see below), the values are more tightly coiled around the center in ’17 and ’18, but spread out noticeably starting in ’19. This suggests that players evolve over time and that a single season of streakiness or consistency doesn’t necessarily define their overall persona at the plate.
When comparing players’ regular season and playoff wOBA at the career level, Team MAC hitters lose about 47 points of wOBA when facing the bright lights, cooler temps, and sturdier pitching of the postseason. Meanwhile, Team Burrito hitters, on average, drop off by just 25 points of wOBA. But as we know, averages can smooth out a lot of extreme scenarios, and given the nature of these two groups, one might expect differences in variability. However, both averages clock a standard deviation around 0.07, meaning the actual results are highly-but-equally variable for both groups of hitters.
The MAC hitter with the most surprising swoon in the playoffs is Nick Markakis, who tops the leaderboard in terms of consistency, while owning a career regular season wOBA of .340 and postseason wOBA of .235. Markakis shows that even the most reliable hitters are liable to underperform in the playoffs. The same is true for their streaky counterparts, but the Burrito crew balances that risk with tremendous upside.
Perhaps the king of the Burrito squad is Hanley Ramirez, who posted a career regular season wOBA of .364 and a postseason wOBA of .434. Right behind him on the leaderboard is Bryce Harper, with a regular season wOBA of .385 and a postseason mark of .415. Harper, alongside his Burrito buddy Schwarber (regular season wOBA .355, postseason wOBA .389), helped the Phillies make deep playoff runs from the wild card position in each of the last two seasons.
With this season’s playoffs rapidly approaching, which active players currently sit atop the frozen food leaderboards?
Both leaderboards boast players on contending teams, but given what we now know about the potential upside for the more burrito-esque players, let’s look at the lineups among the postseason hopefuls that draw most heavily from this archetype. Several contenders have two burritos on the menu:
Atlanta Braves: Marcell Ozuna, Austin Riley
Baltimore Orioles: Colton Cowser, Anthony Santander
Boston Red Sox: Tyler O’Neill, Jarren Duran
Cleveland Guardians: José Ramírez, Josh Naylor
Kansas City Royals: Vinnie Pasquantino, Bobby Witt Jr.
Minnesota Twins: Byron Buxton, Jose Miranda
New York Mets: J.D. Martinez, Jesse Winker
New York Yankees: Aaron Judge, Juan Soto
Philadelphia Phillies: Bryce Harper, Kyle Schwarber
And there were two teams that tied for the lead with three burritos in their lineup, going one step further to stack the odds in their favor come October.
Arizona Diamondbacks: Joc Pederson, Ketel Marte, Christian Walker
Los Angeles Dodgers: Shohei Ohtani, Mookie Betts, Teoscar Hernández
Granted Walker is injured, so the added advantage of an extra burrito may be diminished depending on how he looks when he returns. And I want to be careful not to overemphasize the actual impact of any advantage that might be at play here. Not all burritos are of equal quality, and because we’re talking about an average of 20 points of wOBA applied to a small number of postseason plate appearances, there’s only so much value to be gained in such a short window of time. Overall, it’s still far better to be good than to hope to get lucky with the timing of a hitter’s hot streak. This is the type of analysis that’s mostly fun to observe, as opposed to anything that should factor into decision making. That said, the more chances teams give themselves to catch lightning in a bottle, the more likely they are to catch it.
Which does pose larger questions about roster construction and hitter sequencing. In a future piece, we’ll do a little simulating in an attempt to determine the proper ratio of burritos to mac and cheese, and in what order you should eat them. Until then, stay hungry.
Kiri lives in the PNW while contributing part-time to FanGraphs and working full-time as a data scientist. She spent 5 years working as an analyst for multiple MLB organizations. You can find her on Twitter @technical_K0.
Great analysis, better images #BringBackNotgraphs