The Ball-Strike Machine

So here's what happened: I was watching the end of the Royals-Yankees game on Wednesday … specifically I was watching Joakim Soria suffer again. This has been a tough year for us Soria fans. For four years, he was something really good about the Kansas City Royals. True, I think the Royals should have at least tried in some token way to make him a starter, just to see if it would have worked. True, I think the closer role is absurdly overrated in baseball, and this is especially true for a bad baseball team like the Royals.

Still: He's been wonderful to watch. He has been Kansas City's miniature replica of Mo Rivera -- not unlike the Las Vegas version of the Eiffel Tower, which I love too. Over four years, batters hit .197 against him. He struck out four times as many as he walked. He sliced corners with his fastball and fooled batters with his big, loopy curveball. More than anything, he exuded calm, something the Royals have not really featured in many, many years. It was like after all the madness, all the blunders, all the losing streaks, all the comical and not-so-comical disasters, the Royals could finally go to the bullpen, call Soria, and he would take the mound and every movement he made suggested in powerful ways: "OK, this game's over now. I got it from here."



It hasn't been that way for most of this year. Oh there have been stretches -- he was marvelous in June and July, for instance. He appeared in 20 games, had a 21 to 2 strikeout-to-walk, the league hit .188, he saved all 12 of his chances and won two more. He looked mostly like himself. But before that point, he was so lost that he briefly lost his role as closer. And since July 30, he has reverted back to that lesser version of himself -- he has had only one clean inning in his last eight appearances, the league is hitting .375 or so against him over that time, his ERA approaches 9. It's been tough.

Because of this, I knew -- every Soria fan knew -- Wednesday was going to be a struggle. He came in the ninth inning to protect a two-run lead against the New York Yankees. And as usual, he was also facing the crushing momentum of Royals malfunction -- Kansas City had lost four in a row, eight of nine. The Yankees would send up a series of Hall of Famers or near Hall of Famers. This was going to be a struggle.

Soria managed to get Brett Gardner to fly out after a long battle, but he could not do anything with Derek Jeter, who nostalgically looked like the 1999 model during this series. It was Jeter's fourth hit of the game, a hard line drive to center field that Melky Cabrera had in his glove but could not hold, probably because he seemed intent on diving.*

*Speaking of stunning: Cabrera has had a very good offensive year. But he seems to me, in limited viewing, almost unwatchable as a center fielder.

Curtis Granderson followed with a single of his own. And then agony began. Soria, whose default position is calm, who has always seemed so unflappable, seemed to run out of ideas. He walked Mark Teixeira on four pitches to load the bases. He then went through a seven-pitch at-bat with Robinson Cano where, honestly, he must have thrown three pitches into Cano's wheelhouse … as if he was Jose Cano trying once again to win his son the home run derby. I mean these were 90 or 91 mph fastballs up in the zone, the worst pitches imaginable. Fortunately for Soria, Cano fouled them off. Maybe Soria had just enough subtlety to force those foul balls. More likely, I think, Cano just had a bad at-bat. He eventually hit a sac fly that scored a run. The Royals led by one.

Soria then walked Nick Swisher on four pitches -- one of them a passed ball where it looked like Soria crossed up his young catcher Salvador Perez. Bases loaded again. Two outs. Jorge Posada stepped up. And this was the at-bat that sparked this blog post. Soria's first pitch was way off for a ball. His second looked to be about three or four inches outside -- and the umpire called it a strike. The little ball-strike box they show on TV suggested it might have been even more outside than three or four inches. Soria then threw a nifty little pitch -- MLB defined it as a slider, but it looked more like a very tight curveball -- at the knees on the outside corner for strike two. That was one heck of a pitch.

And then, Soria threw another fastball, this one probably two inches outside. And the umpire rung up Posada, and the Royals won the game.

I was happy, of course, both for Soria and the Royals fans who deserve a little joy. But the idea suddenly struck me -- Soria had probably thrown only one strike in that at-bat, and even that one was RIGHT on the bottom corner. And so I tweeted this:

"Wonder what would happen if they invented a machine that called balls/strikes. Soria would not have K'd Posada is one thing."

The problem with Twitter, as mentioned here hundreds of times, is that a specific thought tends to take on new meaning when you throw it out there naked like that. It almost instantly occurred to me that some people might have thought I was singling out Soria and the Royals -- and sure enough the Twitter responses immediate griped about a missed call the night before, how the Yankees always get the calls, blah blah blah, which wasn't my point at all. I honestly wondered -- and still wonder -- how different would the game be if there was a perfectly accurate ball-strike machine that called the pitches without emotion, without momentum, without prejudice. So to even up the Twitter score, I made what seemed to me an obvious point -- there is another closer who has gotten a few nice ball-strike calls in his day too.

But another might be that Mariano Rivera wouldn't have like a 0.000000048 lifetime ERA.

This may surprise you … but that second tweet did not actually navigate us back to the point. No, now, apparently, I was ripping Mo. And so the responses came roaring in -- how dare you, blasphemy, desecration, sacrilege and so on. I have to admit: I love Mo Rivera, he's one of my all-time favorite pitchers, and I honestly thought Yankees fans knew that he gets favorable strike calls. I didn't think it was even a bad thing. But then, I thought Dr. J fans knew that he was never called for traveling, I thought Broncos fans knew that their offensive linemen got away with Kung Fu Grip holds for decades, I thought Duke fans knew that their teams tended to get more than their share of block-charge calls. But whenever actually wrote such things, I found that the fans actually did not know that or believe that or appreciate anyone saying it. I never learn.

And here is where we finally get to the point. I put it out there on Twitter: OK, fine, who in baseball history would be helped or hurt most by an accurate ball-strike machine? More than 100 responses poured in over the next 45 seconds. Dozens more came in through the night. I just got one as I was typing this paragraph. Almost all of them talked about who would be hurt most. And here was what surprised me.

Almost every single person said "Greg Maddux" or "Tom Glavine."

Almost every one. I mean whenever I throw little questions like this out there, the answers tend to be all over the place. But not this one. There was an occasional "Livan Hernandez" response, referring to that famous game Hernandez pitched against Atlanta with Eric Gregg behind the plate. There was an occasional Mo Rivera thrown in. But as the night progressed, as more and more people responded, it was running 98% Maddux or Glavine.

I want to focus on Maddux -- my favorite ever pitcher -- but I need to talk about Glavine first. Tom Glavine won two Cy Young Awards. He won 305 games. And he left behind a reputation as a great control pitcher, though it's kind of a funny thing what stats tell you: Glavine was not a great control pitcher, assuming that "control pitcher" means a pitcher who doesn't walk people. He only once in his career finished in the Top 5 in fewest walks per nine innings. More to the point, he finished in the Top 5 in MOST walks five different times in his career. He's 12th all time in walks with exactly 1,500. His strikeout to walk ratio is 16th of the 17 who won 300 games since 1901 -- ahead only of Early Wynn. Jamie Moyer doesn't just have a better strikeout-to-walk ratio than Glavine, he has a MUCH better strikeout-to-walk ratio.

In other words, I think Glavine has been wildly misunderstood as a pitcher. Yes, it's true, he seemed to go entire games without throwing actual strikes. We used to talk about that all the time when we watched him pitch. And an enduring image of him might be of an umpire ringing up a batter on a pitch six inches outside. But I think what made Glavine great was the way he imposed his will on hitters. Sure, the umpire might give him a generous call or two or five, but he was so consistent, so relentless, that batters would find themselves swinging at the lousiest pitches or find themselves trying to crush that circle change and basically just getting themselves out.

I do think -- as King Kauffman and Will Carroll tweeted -- that Glavine used the moving strike zone to his advantage, that he was a master at using whatever advantages the umpire gave him*. But I also think that if the strike zone was called exactly to specifications by a bloodless machine, Tom Glavine still would have been a great pitcher. He just would have been a different kind of great pitcher. I have great faith in Glavine's abilities to figure out ways to get hitters out.

*You might remember that for years, Glavine was pretty famous for struggling in first innings and pitching great from that point on. And it is true, even over his lengthy career, that batters had a .354 on-base percentage against him in the first, by far the highest of any inning. People always had theories about that, but I wonder if it was simply that it took Glavine's calculator-like mind an inning to figure out what strike zone the umpire happened to be calling that day. And once he figured it out -- or, if you prefer, once Glavine himself defined the strike zone with his hypnotic pitches -- it was a lot easier for him to get batters out.

To Maddux. I don't think there's really any question that Maddux in his prime was the beneficiary of a wide strike zone. He didn't just achieve this with his name or reputation. He achieved it by expanding the zone with his almost ludicrous command of pitching. He pitched the way a brilliant shell-game operator works the cup and balls -- he showed the umpire a pitch a tenth of an inch outside, then pitch a half inch outside, then a pitch three-quarters of an inch outside, and so on, until he could throw a pitch into Centennial Park and the umpire would think it grazed the corner.

But you have to remember: Maddux worked in a pretty limited world for pitchers. Nobody was calling the high strike in Maddux's prime. And by "high strike," I mean belt buckle was often called a ball -- nobody ever called the letters for a strike. The low strike was also virtually extinct. There really weren't many places for pitchers to throw pitches in the 1990s, which I have always thought was a big reason why offense skyrocketed. So Maddux figured out a way to beat the system. He tried to annex more and more land for home plate on the perimeters. And I think he was often successful.

Maddux was so precise and his pitches had so much movement, I have absolutely no doubt in my mind he would have been every bit as great with a machine umpire. If you gave Maddux the high and low strike? Are you kidding me?

The most interesting thing to me is that when I asked who would get hurt by a machine umpire, people almost unanimously suggested Maddux, Glavine, Rivera, Livan Hernandez -- in a single word: "Pitchers." I think that's incredible -- that means even with all the crazy offense we've watched over the last 15 years, people's gut reaction seems to be that umpires have helped pitchers more? Isn't that kind of nuts? I figured for sure that people would say a machine umpire would have hurt Manny Ramirez or Jason Giambi or Gary Sheffield or Barry Bonds or Mark McGwire or Jeff Bagwell or any one of these amazing hitters who walked a ton and worked favorable counts.

My friend Jim Thome hit almost half his 600 home runs when ahead in the count. No pitcher on earth wanted to mess around with Jim with the count 2-0 or 3-1. But how much different is the situation if that 1-0 or 2-1 pitch is called a strike. And how often did an umpire give a borderline call to Jim Thome because he's Jim Thome?

I think this probably says something about our perceptions of the game. I think most people more or less agree that baseball would be a different game if we could invent a perfect ball-strike calling machine. But would it be a better game? I've put a poll up … you can answer for yourself.