Point/Counterpoint: Clayton Kershaw’s 7th Inning

During the regular season, the Dodgers went 23-4 in games started by Clayton Kershaw, and 20-1 since the beginning of June. Kershaw and the Dodgers started to feel invincible, so, naturally, with Kershaw in the playoffs the Dodgers went 0-2, their year ending on a Tuesday in St. Louis. Plenty of things happened in Game 4 that played a role in determining the outcome, but this tells an awful lot of the story:

Kershaw’s final pitch of 2014 was a home run that turned a two-run lead into a one-run deficit. Under ordinary circumstances, that would just be a thing that happened. However, Kershaw had gone beyond 100 pitches, on short rest, in the playoffs. Despite consensus opinion, the Dodgers did have a bullpen, and a loss meant their season was over. So a tremendous number of people now believe Kershaw shouldn’t have pitched in the seventh, that Don Mattingly hung on a few minutes too long. And, certainly, we here have talked an awful lot about the need to be aggressive with bullpen usage in October. To talk this particular case through, from both sides, I’ve enlisted the help of my own brain. Usually it likes to sit my posts out, but this is a special circumstance.

Point

Six shutout innings, one hit, and 94 pitches? Don’t you have to look at that and say, “he’s given more than enough for a guy on short rest?” Why push it? You know you’ve got Kenley Jansen waiting in there eventually. Just how little can you reasonably trust big-league relievers on a playoff roster? Let’s not exaggerate the Dodgers’ bullpen disaster. They should be able to go a couple innings without allowing a couple runs. That would be an ERA of 9. The Cardinals aren’t some offensive juggernaut.

Counterpoint

Nothing against you, but it’s impossible to read this and not feel like it’s just being reactionary, responding to the results. Hindsight makes every decision a simple one. Listen to yourself, now: six shutout innings, one hit. One hit! Nine strikeouts! You know what Kershaw did in the sixth inning? He struck out Pete Kozma on three pitches. All right, I know, I know. He struck out Matt Carpenter on six pitches. He struck out Randal Grichuk on seven pitches. Maybe that’s a lot of pitches for striking out the side, but Kershaw also didn’t leave anything over the plate, and he finished off Grichuk going 74-94-88. There was absolutely no indication that Kershaw was fighting it. The only indications were that Kershaw was Kershaw, and that guy was probably the best player in the National League, at least.

Point

Kershaw’s sixth inning was great, absolutely, even adjusting for the Kozma factor. But then, we know how a pitcher looks in one inning isn’t very meaningful when it comes to how he’ll look the next. Between innings, there’s a long break, and no one’s ever been able to prove a “hot pitcher”. Couldn’t you also say that maybe Kershaw exerted himself striking out Carpenter and Grichuk on 13 pitches? Every pitch takes a toll, and not every toll is the same. Those were difficult strikeouts. Kershaw could’ve been feeling it.

Counterpoint

Kershaw this year had 19 starts of at least 100 pitches, and five of at least 110. There can be stressful at-bats in the first and second innings, too, but that doesn’t mean you take a pitcher out. There was no sign of coming disaster. Not one. While the short-rest thing was a concern, Kershaw looked every bit like himself, so you treat it like it’s any other ordinary Kershaw start. Why take him out after 94 pitches if he’s cruising?

Point

Don’t you have any faith in fresh relievers? It’s funny — exactly a year earlier, Kershaw started on short rest for the first time. He was good then, too, and the Dodgers removed him after 91 pitches. And they didn’t remove him because they needed a pinch-hitter. They removed him so that Ronald Belisario could face Andrelton Simmons, Elliot Johnson, and Jose Constanza, and they suck. Exactly a year earlier, the Dodgers weren’t afraid to go to the bullpen maybe a batter too early instead of a batter too late.

Counterpoint

And Johnson tripled, and Constanza singled, and-

Point

That can’t be your argument.

Counterpoint

Last year was last year. Circumstances are always different. The start before that one, Kershaw had thrown 124 pitches. In his last inning against the Braves, he generated a pair of line-outs. Look at the Dodgers bullpen. Jansen? Amazing! The rest? What’s the opposite of amazing? The first guy Mattingly had warm up was Pedro Baez. It doesn’t get much better after that. It’s not just about starter with a high pitch count vs. fresh bullpen. It’s about the specific starter vs. the specific fresh bullpen.

Point

But there’s a times-through-order penalty, isn’t there? It exists for everybody. Kershaw’s not immune, and he was going through the Cardinals for the third time in the game. That means something. All the hitters had seen him twice before on the day. You have to adjust your expectations for Kershaw down. Or up, I guess, if you’re talking about batting average or wOBA. This is something that’s so easy to forget, but it’s real.

Counterpoint

Totally, totally, that’s true. Pitchers get worse as they get deeper in games. But let’s look at Kershaw specifically. Over his career, the first time through, he’s posted a K% – BB% of 22%, with a .097 ISO. The second time through, he’s gone 18% and .105. That’s a little worse! But combine the numbers for the third time through and the fourth time through. We get 17% and .095. It’s the same, basically. Kershaw, himself, hasn’t demonstrated much of an issue after the first time through the order. That doesn’t tell you about Kershaw on Tuesday in particular, but with the quality of his pitches, and with his deception, is it such a surprise? Carpenter and Grichuk didn’t look too comfortable.

Point

But then there’s the part about being on short rest. We know that, on average, there’s a short-rest penalty, and while we don’t know exactly how it shows up, we know more runs score and we know pitchers don’t go as deep. Doesn’t it make sense that pitchers on short rest might fatigue a little earlier? Like, say, around 90 pitches, instead of 100 or 110? This can’t be dismissed. This was kind of the very biggest concern.

Counterpoint

But we don’t know. We don’t know anything about how short rest affected Kershaw in Game 4. It sure seemed like he was himself. His pitches weren’t any slower or worse. His sixth inning was great, leading up to 94 pitches. He said he felt great at the time, and while you can take that with a grain of salt, he also said he felt great afterward, physically speaking. Kershaw hasn’t referenced any kind of fatigue. If anything, I think maybe the short-rest effect was going to show up later, in the next start or the one after that. On the day, Kershaw was fantastic until he wasn’t.

Point

Wouldn’t you rather be too conservative than too aggressive? If you don’t know what the effect is going to be, wouldn’t you rather go with normal relievers on normal rest? It was Kershaw’s third time through the order, and he was on short rest, and while you don’t know exactly what that would’ve meant, it probably wouldn’t have meant anything *good* for his endurance. It’s either no effect or some effect, right? So that averages out to some effect. With the bullpen, you can just play matchups to get you through until it’s Jansen time. You’ve got Dan Haren to soak up innings in the event of an emergency. Don’t try to pretend like the bullpen was an automatic loss.

Counterpoint

Let’s assign something of a short-rest penalty. Let’s also assign something of a times-through-order penalty. Now, think about Kershaw’s foundation. He’s an elite starting pitcher — probably the best in baseball. So that’s his starting point. Then you knock him down a little bit. Then you knock him down a little bit more. What do you have left? Probably still a tremendous pitcher, unless you’re super aggressive with your dings. What are the expectations, with Kershaw at X% vs. relievers at 100%? Kershaw this year allowed a .521 OPS. Jansen this year allowed a .610 OPS. And he’s the stopper. Kershaw is so, so good.

Point

He is really good.

Counterpoint

So good!

Point

Look at how the seventh went, though. Matt Holliday immediately singled. Jhonny Peralta immediately singled. That indicates danger. One hit through six. Two hits in the seventh through two hitters. Mattingly’s plan was apparently to get Kershaw through three hitters. It took almost zero seconds for it to not work out.

Counterpoint

On the other hand, the Holliday single bounced off Dee Gordon‘s glove, and the Peralta singled bounced off Hanley Ramirez’s glove. Holliday’s was a groundball up the middle. Peralta’s was a soft liner. The contact was fine, but not great, and both plays could’ve been outs with just the slightest little adjustment.

Point

I don’t care about how close they were to being outs. A home run is a centimeter on the bat away from being a pop-up. Look at the pitches. The curveball that Holliday hit missed to the opposite side of the plate.

Counterpoint

The fastball that Peralta hit was just about spot on, on the inside edge.

Point

…and it was set up by a bad first-pitch slider in the dirt, and a second-pitch slider that hung up. Peralta hit it hard, just foul. Kershaw was missing spots. Was that his fatigue? If so, how do you let him continue? Why wasn’t there even a lefty throwing in the bullpen?

Counterpoint

So Matt Adams was the lefty that Kershaw had to face with two on and none out. Mattingly’s bullpen alternatives were J.P. Howell and Scott Elbert. I don’t care what you think about Howell and Elbert — even Kershaw at like 75% is a hell of a bet in that situation. Adams has a career .243 wOBA against southpaws, with nine times as many strikeouts as walks. Kershaw’s allowed a career .251 wOBA to lefties, with five times as many strikeouts as walks. If you plug some numbers into the odds-ratio formula, you get an expected matchup wOBA of [fart noises]. Kershaw is the lefty specialist the Dodgers wanted there. Even if you disagreed with Kershaw facing the first two righties in the inning, he did that, so you might as well leave him in for the lefty. He ought to be able to get the lefty. And then he was going to get removed no matter what.

Point

I’m starting to think maybe there wasn’t an obvious decision to be made.

Counterpoint

And there wasn’t. This wasn’t like Matt Williams leaving Tyler Clippard in the bullpen, or Mattingly pinch-running with Yasiel Puig while pinch-hitting with Justin Turner. There were reasonable arguments for removing Kershaw. There were reasonable arguments for sticking with Kershaw. Absolutely nothing about the decision was easy. And here’s the little secret: if you’re almost equally torn in two opposite directions, that means that neither choice is a bad one. If you can see leaving Kershaw in, and if you can see taking Kershaw out, that means there’s probably very little change in the odds no matter what you elect to do. So it seems like a really difficult thing to sort out, but in reality it’s simple. If you don’t know what to do, it doesn’t matter that much what you do. Either way, the numbers hardly budge.

Point

That’s not very fun to debate, of course. It kind of shuts the conversation down. People love to second-guess. People live to second-guess. And I think it’s always reasonable to reflect on decisions that get made, to try to at least learn from them for the future. It helps to see how managers think, and managers can also learn from their own experiences. No sense in always appealing to authority. Refer to, again, the weird-ass Puig/Turner thing. Is he hurt or not! If he’s hurt, why is he pinch-running!

Counterpoint

It certainly doesn’t make for compelling content. But I think this is where the issue is: after the fact, people want to believe that there is a right decision and a wrong decision. Like, when a manager is faced with a choice, and he has two doors, then behind one of the doors is success and behind the other door is failure. For Mattingly, on Tuesday, he opened one door and he was left to deal with failure. People want to believe success was only one door away, so, why didn’t he choose that door instead? But that’s not how this goes. With Kershaw, odds were the Dodgers would get out of the seventh still ahead. With relievers, odds were the Dodgers would get out of the seventh still ahead. The problem with life is we get a sample size of 1. Great to live, sucks to analyze.

Point

Now I feel like we’ve kind of arrived at the same place. Mattingly had the toughest decision to make in the world, which also meant it was a pretty simple decision to make. So what happens between all the manager decisions and the end of the game?

Counterpoint

Baseball happens.

Point

Like Clayton Kershaw allowing a home run on a curveball to a lefty for the first time in his big-league career.

Counterpoint

It’s all so beautiful and stupid.