In baseball’s October blame game, direct your ire toward a person, not the ‘analytics’ bogeyman

The baseball year, like the calendar year, has seasons. There’s projection season, from December to March, when the industry tries to foresee and maximize the future. There’s assessment season, from April to July, when we figure out which teams are contenders and what they need at the trade deadline. There’s credit season — August and September — as teams surge or fade to lock in postseason berths and win totals, as players sew up MVP hardware and black ink and record-book positioning.

Then there’s October. And if we’re being honest, the short spurts and short fuses of playoff baseball tend to wind up dominated by one thing: Blame.

Call it the winter of baseball’s discontent, a holiday from reason and long-term thinking. From the first chill until Macy’s starts prepping for the parade, Mercury is in retrograde over the ballparks of America. Playoff baseball has always heightened the senses. Deciding the fate of a season in bursts of two-to-seven games isn’t scientific or rigorous, but it is highly entertaining. So we zoom in, sit up and scrutinize. It’s a monthlong Super Bowl of strategic second-guessing.

In a sport consumed by data and its applications, trying to grapple with the binary, all or nothing, n=1 moments of postseason baseball can be disorienting, like looking for the meaning of a love story in an individual pixel on your TV screen.

Even as MLB teams have collectively moved into a fairly uniform era of data-informed, human-implemented management — beyond the false dichotomies of stats vs. scouts or jocks vs. nerds — the stakes of pivotal playoff moments sometimes make otherwise reasonable minds revert to howling at the moon. Or, more specifically, at “analytics.”

Let’s get this out of the way: “Analytics” is a catch-all term for “data and information that influence how sports teams operate.” In practice, however, it often connotes “data and information that influence sports in a way I don’t like, for reasons I don’t want to explain.” When the same advancements lead to a young pitcher dramatically improving his slider, a veteran hitter vaulting to stardom or a team soaring to its most successful season ever, you don’t typically see those positive developments attributed to one vague buzzword. You’ll instead see the credit distributed to the people who utilized the available information to make good things happen.

When bad outcomes must be explained, though? There’s a one-word answer. Spreading through MLB managers like Havana Syndrome ravaging spies, the scourge of “analytics” has been deemed responsible for such nefarious deeds as robbing the world of a no-hitter by pulling Atlanta Braves pitcher Ian Anderson, kneecapping the Tampa Bay Rays in the 2020 World Series by removing Blake Snell too soon and derailing the 2019 Los Angeles Dodgers by leaving Clayton Kershaw in too long.

Most recently, some combination of numbers and the Toronto Blue Jays front office has been accused of forcing manager John Schneider to remove José Berríos after three-plus innings of scoreless ball and turn to starter-turned-playoff-reliever Yusei Kikuchi in Game 2 of the wild-card series against the Minnesota Twins. The Athletic’s Ken Rosenthal held up that decision — which quickly backfired and contributed to a season-ending Blue Jays loss in which they also, importantly, failed to score — as evidence of some vast rot in the game, the loss of “feel” and “trust” in favor of “bloodless, dispassionate decision-making.”

In trying to explain his disgust, Rosenthal stumbles into revealing the subtext of the misdirected furor often vocalized as “analytics.” He notes that Yankees fans are angry at GM Brian Cashman for “analytically driven decision-making … that sometimes defies common sense,” but acknowledges “Cashman’s approach would be fine if the Yankees were using the numbers as deftly as, say, the Dodgers.” He praises the Milwaukee Brewers and manager Craig Counsell for “striking the right balance between the objective and subjective” and cites Counsell’s decision to ride starting pitcher Freddy Peralta for a conventional number of innings in … a wild-card Game 2 loss to the Arizona Diamondbacks that ended the Brewers’ season.

I don’t mean to counter Rosenthal’s argument so much as unpack it in hopes of finding more stable footing for discussions amid Blame Season 2023 and the almost inevitable witching hour of Dave Roberts piloting an unsettled Dodgers pitching staff.

The point Rosenthal, a far more experienced and connected observer than me, is making is important to understand because I think he’s honestly reflecting the feelings of a huge swath of fans and perhaps even people inside the game. He says he’s seeking accountability, but that doesn’t seem quite right. Clearly, Toronto’s move was a bad decision. Virtually everyone watching agreed upon that instantly.

And the blame is easy enough to parcel out: Schneider and the Blue Jays front office for the plan, Schneider for following said plan despite new circumstances, Kikuchi for pitching worse than he’s capable of and the rest of the Blue Jays for failing to score a single run. No, the issue isn’t with the outcome (again, the Brewers also lost) or even with the logic.

It’s an aesthetic objection masquerading as a strategic or even moral one.

Styles, they say, make fights. Tracing a manager’s decision not to his own free will but to the general usage of data alludes to the fear that there is only one style — and, thus, no more fight. That’s how you wind up railing, 1984-style, against a dystopian vision of the present.

The trouble is, it’s not an accurate vision of the present. The bullpen phone is not an automated feed of instructions based on AI-crunched, in-game projections. Someone picks it up and summons a reliever. Catchers are not relaying algorithmic recommendations to the pitchers via PitchCom. They are making decisions — often through the unquantifiable feel for a hitter’s swings — and asking pitchers to trust their judgment.

Whatever decline in the diversity of macro baseball thought you want to pin on the rising tide of data, technology and information, you’re probably correct. Reams of knowledge that became indisputable — mostly that slugging and strikeouts are the most powerful, most valuable forces in the game — led teams to optimize for a style of play that suffered, in entertainment value, compared to many earlier versions of the sport. Sabermetric leader Theo Epstein has said as much. MLB has acknowledged as much, and with new rules this season, it took action to at least begin the process of course-correcting.

The same criticism does not stick on the micro level. Teams with similar resourcing and leadership regularly steer toward diametrically opposed paths — with wildly varying results. You can try to make the case that the successful ones emphasize the human touch over the numbers, but you’d be speaking nonsense. The best baseball teams are the ones that help players be the best versions of themselves. Most of them do that by studying data and information, and then taking care to guide players in whatever ways work best for them. There is art in effectively leveraging the science, but there’s no glory in ignoring it.

What makes the playoffs so tense, so chaotic, is the way they pepper baseball’s best teams with situations that can’t be reduced to science. There’s no trial and error. There’s only the trial; an error could mean you’re out. And it’s OK to evaluate that error as a human one, giving credence to both the intention and the information behind it.

This is part of what Schneider said after that Game 2 loss when asked about the decision to pull Berríos: “I think that when you’re so diligent with your work and you trust the people that you’re working with and the people that you’re kind of going to battle with, both on the field and off, you just try to make the best decision that you can for the guys that are on the field to win.”

In this month of legacy-forming snap decisions, managers of similar lineage make inconsistent — and interesting — decisions all the time. Hell, managers of different lineages evolve toward making similar decisions. No matter how much one believes modern front offices alter dugout strategies, stripping everyone involved of credit or shame for the resulting triumph or defeat is tantamount to misconstruing the heart-pounding reveals as a sterile simulation.

Yes, there’s a common framework for understanding and analyzing the game now. Yes, there are numbers and metrics that might require some explanation to the uninitiated. But if you provide 10 smart people with access to the same information and ask how they would deploy, say, the Dodgers’ postseason pitching staff in 2023, you’re probably going to get at least eight different answers.

If Roberts and president of baseball operations Andrew Friedman and pitching coach Mark Prior and Co. — architects of a 100-win team — come up with one that falls short against the 84-win Diamondbacks this week, we can examine their decisions and reckon with the unknowables of a 95 mph fastball smashing into a wooden bat. We can highlight their mistakes (within the context that this is an entertainment product) while accepting both their rationale and their humanity. We can analyze and criticize the game as we see it without disavowing the whole enterprise for failing to resemble the actual year 1984.

Blame the manager. Blame the GM. Blame the players. Don’t blame the mere existence of information — which is in use only because so many people care so much about the events at hand.

The fact that there are more people (and database queries) involved in the lead-up to those key moments now — and that some of the styles are extensions of a GM sitting in a suite — doesn’t make them any less thrilling as collisions of tactical ingenuity and athletic execution.

There are plenty of fascinating, intellectually earnest fights to have about postseason baseball. Let’s not waste our punches on a ghost.

Leave a Comment