The Only Rule Is It Has to Work

by Ben Lindbergh

  A few minutes later, I get my answer. Captain sticks his head out and asks me to come back in. “You can’t do that,” Yoshi says when the door is closed. “It’s not your job to talk to players.” I’m still so surprised that I can only stammer something about how I’ve been talking to players all season. “You know what I mean,” he says. “You’re not stupid. You just don’t know how to behave.”

  It’s true that on a typical team, posing questions to players would be the exclusive domain of the manager. But on a typical team, I wouldn’t be signing players, picking them up at the airport, playing Super Smash Bros. with them, giving them dap in the dugout, filming their games, and showing them their video. The point of my presence, and Sam’s presence, is that this isn’t a typical team. To this point, I’ve been Taylor’s contact with the Stompers, the guy he texts when he wants to know what time to get to the clubhouse, or who has the key to the Arnold Field entrance, or whether he can come over while he waits for his host family to take him home. It doesn’t seem like a stretch for me to ask how he feels about playing left field. Moreover, I was just trying to be helpful. I didn’t care who talked to the players, and I asked Yoshi which way he preferred. This seems like a disproportionate response, as if I’d walked up to the whiteboard without talking to anyone and switched their assignments myself.

  But it’s not disproportionate, because it’s not a response to a single incident. He also brings up my impertinent in-game pitcher advice. “I’m here to offer information and recommend moves,” I say. “How would you like me to do that?” Captain butts in to tell me how he thinks I should treat Yoshi, which sounds like an aristocrat telling a debutante how to approach the queen at cotillion: Walk forward when announced, curtsy, and don’t speak unless spoken to. “Sam knows what to do,” Yoshi says. Of course he prefers Sam, since Sam wants to take it so slow the season might end before anyone knows why we were here.

  If I had an attack mode, this would be a good time to enter it. Yoshi is marking his territory, so I should mark mine. Maybe this is the time to push, to point out that in a very real sense I outrank him. But the prospect of turning Yoshi against me for good, so soon after resolving the conflict with Feh, is too terrible to risk. I can’t take another six weeks of cold war, so I try to be conciliatory. I say that I meant no offense, that I respect his opinion, and that I hope we can collaborate. We shake hands. Hurley starts in right and Taylor plays left.

  The drama subsides for six days, until another arrival forces us back into the ring. Theo tells Brennan Metzger, Matt Hibbert’s outfielder friend from Long Beach, that he can come work out with the team, and once he takes fungoes everyone wants him on the roster. He’s a little smaller than Eads, but much more athletic: In the minors with the Giants, who took him in the twenty-second round of the 2012 draft, he played four positions, stole a few bases, and took enough walks to offset some of the damage done by his low batting average and paltry power. He reeks of recent affiliated experience, a perfume that’s intoxicating to Yoshi. We sign him, but we don’t have an empty place to play him.

  After batting practice, Sam, Theo, Yoshi, and I walk down the right-field line, out of earshot of anyone else. I’d hoped that Taylor’s doubles streak would have endeared him to the doubters, but Yoshi wants to bench him to make room for Metzger. Sam and I strenuously object. This is a Tuesday, and the team was off Monday, so I say I’m concerned that we’ll disrupt Taylor’s rhythm if he doesn’t start until Wednesday.

  Yoshi complains about Taylor’s terrible defense and terrible reaction to his first dropped-ball blunder. Sam says he told Taylor in the dugout not to get too down. Yoshi gives him the same “don’t talk to players” scolding I’ve already received, and Sam says it was a human reaction, that he was just trying to comfort a kid who clearly felt awful. Yoshi says that isn’t his job. Now that Sam is trying to give Yoshi orders, he’s apparently no longer the model of preferred front-office behavior.

  “Hurley and Hibbert have been here all year,” Yoshi says. “How can I tell them that they’re not going to play?”

  “What does it say to Taylor if right after he gets going we bench him and bring in another guy?” I ask. Sam points out that Hurley and Hibbert have hardly had any days off. They were already due for a rest.

  “Sam, you don’t understand,” Yoshi says. “You don’t know baseball.” He punctuates each sentence with a chest jab—not the friendly kind, but a display of dominance designed to put Sam in his place.

  Sam’s stiffening posture says he’s had enough jabs. Nor is he happy about being told he doesn’t understand baseball. Yoshi says he’s already implied that Hurley and Hibbert will play, and that if he flip-flops now, after talking to us, it will undermine his authority. “Blame it on us,” Sam says. “Make us the bad guys.” Yoshi scoffs, exasperated by our ignorance of clubhouse dynamics. Sam sighs and plays his trump card: “Taylor isn’t losing his job, so I guess we won’t sign Brennan.” Yoshi stares back in shock.

  After that, we reach an agreement: Taylor will get this game off, but Yoshi will tell Hurley and Hibbert, today, that each of them will take a different day off this week—and Brennan will join the team. We’re all slightly dissatisfied with the outcome, the sign of a successful compromise. The argument got heated, but it led us to a productive place. And unlike Feh, Yoshi didn’t walk away.

  In June, during one of our earliest, daydreaming discussions about replacing Feh with Yoshi, Theo pointed out that Sam and I were probably overestimating how much better life with Yoshi would be. “Yoshi is probably 95 percent the same as Feh,” Theo said. “I’m probably 92 percent the same.”

  “Ben and I are probably about 87 percent the same,” Sam said. “It’s good DNA.”

  Eight imaginary percentage points doesn’t sound like that big a barrier. We just have to work on whittling it away.

  12

  EVIDENCES

  Memo to: Yoshi et al.

  From: Sam

  There’s been a lot of evidence that, in the major leagues, pitchers perform significantly worse the third time through the batting order. This was first reported in 2004 by the noted sabermetricians Tom Tango (a consultant to the Cubs) and Mitchel Lichtman (a consultant to many teams, most notably the Cardinals). The so-called times-through-the-order penalty is likely a combination of two factors: pitcher fatigue, and batters’ increasing familiarity with the pitcher. This is a brief rundown of the evidence for this, factors to consider when we make our in-game decisions, and whether the same thing is true for Pacific Association pitchers.

  Evidence

  First, the evidence at the major league level, where we have the most data and the cleanest data. These are the splits for all batters against starting pitchers this year, for each time they’ve seen the pitcher in a game:

  • First time: .248/.305/.388, .693 OPS

  • Second time: .261/.317/.405, .722 OPS

  • Third time: .268/.326/.434, .760 OPS

  The difference primarily comes in lowered strikeout rates. Pitchers strike out 21 percent of batters the first time they see them, then 19 percent the second time, then 17 percent the third. Walk rates and home runs also go up, mostly from the second time to the third. Here are the pitchers’ FIPs each time through:

  • First time: 3.74

  • Second time: 3.94

  • Third time: 4.54

  FIP is fielding-independent pitching; it creates an ERA-like number based on a pitcher’s strikeouts, home runs, and walks, and it “predicts” what the pitcher’s ERA would be. As with ERA, a FIP below 3.00 is very good, above 4.00 is bad, 2.00 is elite.

  In other words, an average pitcher at the start of the game turns into a fringy pitcher by the end. A no. 1 starter becomes a no. 3, and a no. 3 becomes a no. 5, and a no. 5 becomes outright awful.

  There are MLB teams who are using this knowledge to make their pitching decisions. The Tampa Bay Rays in particular have taken to pulling some starting pitchers once they’ve faced 18 batters. Their starter
s have faced the fewest batters per game of any team in the American League. The Kansas City Royals are second. The Rays and Royals have something else in common: They also have the two best pitching staffs in the American League this year.

  We don’t think this is a coincidence.

  Pacific Association

  But would it work in the Pacific Association? To answer this, let’s see whether the same thing happens to our pitchers.

  These are the team’s strikeout and walk rates each trip through:

  • First: 19.5%, 7.3%

  • Second: 17.1%, 7.4%

  • Third: 17.4%, 11.1%

  As you see, our pitchers strike out fewer and walk more the deeper into a game they get. In five of our nine losses, it has been this very trip through the order in which the game has changed from a Stompers lead to a Stompers deficit.

  Loss 1: Leading 2-1 in the fifth; starter (Schwieger) allowed two in the fifth. Both runs driven in by hitters seeing him for the third time.

  Loss 3: Leading 1-0 in the sixth; starter (Conley) allowed two in the sixth. Both runs driven in by hitters seeing him for the third time.

  Loss 4: Leading 4-3 in the sixth; stater (Walker) allowed four in the sixth. All batters in the inning saw him for the third time.

  Loss 5: Leading 4-2 in the seventh; starter (Schwieger) allowed one in the seventh. The run and the RBI were both by hitters seeing him for the third time.

  Loss 7: Leading 2-0 in the fifth. Starter (Walker) allows two in the fifth, two in the sixth. All runs driven in by batters seeing him a third time.

  Challenges

  The most obvious challenge is that starting pitchers don’t want to come out of games, particularly before completing five innings when a potential W is on the line. It’s up to the pitching coach or the manager to assess how disruptive it would be; however, at this level pitchers generally don’t have the same set patterns or the same financial incentives that a major leaguer does. Most of our pitchers have pitched in various roles in their careers, and should be accustomed to adjusting their routines depending on where they are and what the team needs. The value of a win to their career seems very low—teams in higher indy leagues would, I imagine, be much more interested in the pitchers’ ERAs and strikeout totals than win totals, especially because so many pitchers are shuffling between leagues and because league schedules are of differing lengths; there is no equivalent of a 20-game winner at this level. Affiliated-ball scouts are going to be entirely uninterested in a pitcher’s win totals.

  Regardless, in close games the benefit to using our pitchers in shorter stints would potentially be great.

  * * *

  Forty-one months before all of this started, my wife and I put our eight-month-old daughter to sleep on a cold December night, poured champagne into paper cups, and went outside to walk tight laps in front of our Southern California home. We talked about our favorite moments of 2011, and declared our resolutions for 2012. This was a lifetime ago; I didn’t work for Baseball Prospectus yet, I hadn’t moved to the Bay Area, I’d never heard of the Stompers. My resolution was inspired mainly by my bosses, whom I hated, and whom I had increasingly found myself arguing with. These arguments never made them like me more, I had found. Further, I never won these arguments. (Even though I was always right!) So my resolution was simple: Never try to win an argument.

  It wasn’t just my bosses. I’d grown tired of arguing, with my wife, with people on the Internet, with friends who thought that the Notorious B.I.G. was a better rapper than Ghostface Killah. Once I started trying to win an argument, I found myself rotating every fact to suit my position. This was true of the facts that came out of my mouth and also of facts that went into my ears, which I heard only deeply enough to deflate or reposition. Conversation became an exercise in bullshit.

  A few months after this resolution, I read an article in the New Yorker that made me intensely happy to be argument-free. Its premise, based on the work of political scientists, was that the worst thing a president can do to advance his positions is to state them; as soon as he does, a huge number of people will position themselves in opposition, and they will lose the ability to be swayed by any contradictory evidence.

  That’s in politics, of course, where partisanship separates people into intractable opposition against each other. But after three years of writing about baseball and two years on Twitter I had come to feel that nearly everything was a partisan struggle—at least, once it had turned into an argument that people attached themselves to. This was the ugly part of the stats-vs.-tradition debate in baseball: Rather than a conversation about the best way to make baseball decisions, it had become an argument, in which it increasingly felt as if the purpose was to score points by humiliating one’s opponent. So far as I could tell, nobody had ever changed his mind about anything in an argument. Dale Carnegie was way ahead of me: “If you tell them they are wrong,” he wrote, “do you make them want to agree with you? Never! For you have struck a direct blow at their intelligence, judgment, pride and self-respect. That will make them want to strike back. But it will never make them want to change their minds. If you are going to prove anything, do it so subtly that no one will feel that you are doing it.”

  Indeed, I’d become convinced that the only way our minds are changed is by slow absorption, the feeling that other people we respect all believe different things than we do. The best argument is, essentially, peer pressure.

  Of course, this is a problem when taking over a baseball team filled with players and coaches who need to be persuaded. Ben and I (and Theo) clashed over whether we should rule as tyrants. The deeper into the season we went, the more I appreciated Ben’s chagrin about my reluctance to dictate terms. But I was less interested in the question of what would happen if we quit sacrifice bunting and more interested in the question of whether we could persuade people to do it because it’s good and right and rational. Hence the speech in spring training: We’re going to go slow, so be patient with us. We wanted to win them over, not overwhelm them or argue them into submission.

  But that came with risks.

  Back before this started, when our roster was a blank whiteboard, we weren’t worried about losing; everybody loses. We weren’t worried about our ideas failing; that’s how science happens. We had one worry, which was that two months into the season a friend or colleague or podcast listener would come out to watch the Stompers and it would look like … just baseball. That they’d watch a few innings and ask us, with a bit of disappointment, which part was ours. That it wouldn’t be obvious. Or even that there wouldn’t be a part that was ours at all.

  A few hours after Yoshi poked my chest and asked why I was so hostile—when I was trying so hard not to argue at all—I watch Paul Hvozdovic start against San Rafael, and I struggle to answer the question even to myself. Which part is ours? I scribble a list of ways the Stompers are different because Ben and I had been born: There are scouting reports in the dugout, and we’ve been able to provide players with the sort of pregame preparation that only major leaguers typically have access to. There are a handful of players—some who were essentially retired before we called them—who have become stars in this little Galápagos, thanks to our spreadsheet and faith. That spreadsheet has turned out to be surprisingly progressive and has helped make baseball (and cultural) history. We have the first Japanese-born professional manager in American baseball, and he is (in theory) a statistically enlightened one. There are some defensive shifts, otherwise unheard of in this league. These are all things we’re proud of—but they are more about helping the team reach basic levels of competence than they are helping the team to be visionary.

  We’d dreamed of pushing baseball twenty years forward. We’d envisioned, for instance, a system of calling pitches from the dugout using a computerized random-number generator. The way we (and Sean Conroy) see it, most pitchers are fairly predictable; hitters have a pretty good idea what’s coming because the game theory of pitch sequencing (I know he’s looki
ng for this, so I’ll throw him this) has become predictable. Rather than try to outsmart the hitter using this game theory—and following the same predictable routine—we wanted to have a computer select pitches using a random-number generator. It would be extremely difficult to set up. For one thing, just because the pitch-calling would be random doesn’t mean that we would want the pitcher to throw all of his pitches at identical rates. Sean’s slider, for instance, is much better than his changeup, so he should throw his slider about ten times as often, and in certain counts (where a swinging strike is more valuable, or a pitch in the dirt less dangerous) maybe even more than that. But even if the slider is going to be 70 percent of his pitches, the decision about when to throw the exceptions should be decided using complete unpredictability. (I, too, understand how much this paragraph makes me sound like a philosophizing stoner.) In short: We would have had to figure out how often to throw each pitch, for each pitcher, in each game situation, and how to relay that call to the catcher and/or the pitcher in just a few seconds, and how to convince the pitcher to throw a pitch that he doesn’t want to throw—the basic concept of this whole thing being that sometimes the wrong pitch is the right pitch, especially if the wrong pitch makes future right pitches even more right. (Whooooa.)

  You are not reading about the success of our random-number-generated pitch-calling, though, because of course we hadn’t implemented it—hadn’t even gotten close. We also hadn’t implemented our plan for making pitching changes based on arm-fatigue metrics that would be relayed to our iPhones via a futuristic wearable sensor called the mThrow. In the lead-up to spring training, Ben envisioned us as Ivan Drago’s training team from Rocky IV, studying readouts and graphs as our sensor-equipped players pummeled their opponents. We talked to a lengthy list of cutting-edge companies—brain-training, eye-training, bat-tracking, motion-sensing, injury-preventing—but have little to show for it, only partly because most of this stuff is still in its infancy and very much unsettled science at the dawn of the wearable-technology era. We hadn’t reimagined batting practice in a way that would give our guys more focused preparation while conserving their energy for the game. There are no naps in our clubhouse, and our guys are eating boring peanut butter on white bread before games, and if they are bunting less than the typical team and stealing bases at a higher success rate and drawing more walks and hit-by-pitches (all things we have encouraged), it’s not because of us. It’s because of them. Our most direct contributions have been limited to a pretty phenomenal gumbo night prepared by Kortney from our scouting staff, a Culinary Institute of America graduate who specializes in Cajun food; Sean’s ice bath during his TV interviews; and the one time we used PITCHf/x to determine balls and strikes in an exhibition game, which went well except that the players who made the calls from our computer made deliberate mistakes to mess with their friends on the field. Our dreams have become small. Our lives feel small.