"He's either going to win it or lose it"

Craig already had his say about the following bizarre quote from Royals manager Ned Yost -- who pulled starter James Shields in the ninth inning Monday with the Royals leading Chicago 1-0, replaced him with closer Greg Holland and watched the Royals lose in extra innings 2-1. But I think I have something a little bit different to say about it.

First, here’s the quote:

“Everybody has their job to do and Shields had done his. He threw eight shutout innings. It was a one-run game. The runs make all the difference. If it was a two-run or a three-run lead, yeah. But in a one-run game, you send him out he’s either going to win it or lose it. You let the closer go out and try to do his job.”

Yost is not always the best at expressing his decision-making process. For instance, I’m not at all sure what the English translation is for “In a one-run game, you send him out he’s either going to win it or lose it.” Wouldn’t that be true for, you know, anybody? And, actually, is it even true? He could give up one run and tie it -- might not win or lose it at all.

I think -- think -- he’s saying that in a one-run game like this there’s no margin for error, so, um, I don’t know, maybe you can’t trust the starter to protect a one-run lead because, you know, he did his job already and he can only win or lose the game, or something. It’s the best conclusion I can come up with.

I should say … I’m not second-guessing Yost’s DECISION here. That’s not fair, especially with it failing. It could have worked. I’m second-guessing his EXPLANATION for the decision. If he had said: “I thought Shields was showing signs of fading in the eighth inning,” hey, OK, that’s the manager’s job to notice that stuff. If he had said, “James had already thrown 102 pitches, and I really didn’t want him throwing 115 to 120 this early in the season in somewhat cold weather,” once more, hey, that’s the manager’s job to look out for his players over a long season.

But this explanation seems utterly baffling. And with the Royals off to a really good start, the choices Ned Yost makes (at least for now) will be watched a lot more closely. He will definitely need to come up with better explanations that this cockamamie thing.

Consider:

1. James Shields and Cliff Lee have started more ninth innings than any pitchers in baseball from 2010-2012. It’s only 18 games, a small sample size, but it’s all we have to work with. Batters have hit .174 against Shields in the ninth inning over those three seasons. They have slugged .217. He has not allowed a home run. So to say that pitching eight shutout innings is his job, well, I think a short but sweet history would suggest otherwise.

2. Shields had allowed two White Sox hits, both singles, and was poised to face Jeff Keppinger, Alex Rios and Adam Dunn who had gone 0-8 with two strikeouts, with a walk (to Dunn) and a moderately hit routine fly ball being the highlights.

3. Is Yost really saying that if the Royals had been up 2-0, rather than 1-0, he would have let Shields start the ninth? Because, um, that seems to be really, really unsound thinking.

4. Greg Holland, a very talented young pitcher with a great future, has not yet proven himself to be Mariano Rivera.

And, most of all there’s this: Here are some fun numbers for you. I looked from 2010 to 2012 at starters who pitched in the ninth inning. It’s not a huge sample as you might guess -- complete games are way, way down. Plus, the numbers are kind of hard to harness -- I’m sure smarter people can do a lot more with this. Still, using Baseball Reference’s excellent splits feature, I was able to break it down at least a little bit.

Then, over the same time frame, I also looked at how the top relievers have pitched in the ninth inning -- I basically took the 100 relievers with the most ninth innings pitched.

Here’s what the numbers say according to the muddled spreadsheet I put together:

Batting average against starters in the ninth: .224

Batting average against relievers in the ninth: .225

Slugging percentage against starters in the ninth: .340

Slugging percentage against relievers in the ninth: .345

Walks per nine innings for starters in the ninth: 1.79

Walks per nine innings for relievers in the ninth: 3.16

It’s difficult to figure out ERA or runs per nine innings because pitchers come in and out so often during the ninth, leaving behind base runners, allowing inherited runners to score, etc. But best I can tell, starters also have a better ERA and allow fewer runs per nine than relievers.

Now, there’s all kinds of noise in this little study, all sorts of nonsense, all sorts of small-sample size gibberish, and I don’t want to make too much of it, and I don’t want to make it sound like I’m saying starters are better in the ninth than relievers. I’m not. I have no idea.

I’m just saying that since 2010, starters who had pitched well enough to start the ninth inning tended to pitch well in the ninth inning too. I don’t think that’s a surprise. It seems almost certain to me that the Royals best chance to win on Monday was to send Shields back out for the ninth inning. He had dominated for eight innings, he had only thrown 102 pitches, to an outside viewpoint he looked good in the eighth, I think it’s highly unlikely that bringing in a new reliever who had pitched the day before was the team’s best shot at victory.

Again, I’m not saying the decision was wrong. I’m saying that I would like to know why Ned Yost did it …and his explanation does not clarify things at all. If he did it for health reasons, because he noticed something, or for sound reasons he would rather not make public, hey, that’s part of the game too.

But he did not offer any of that in his explanation. And if he did it because he believes in push-button managing with a starter having his job, a closer having his job, a setup man having his job, a lefty specialist having his job, well, that’s not great. The Royals have a pretty good team here. Ned Yost has gotta raise his game.