Wednesday, June 15, 2011

Tragedy of Errors

Look, bitching about the manger is one of the oldest traditions in sports and it can frequently mark you as a mouth-breathing, talk-radio show calling ignoramus, and I understand that, but the last two games from Ron Roenicke have just been complete debacles. Train wrecks. Ron Roenicke didn't see the iceberg. Ron Roenicke decided to go with Hydrogen over Helium.

And what makes it all the more maddening is that just a few days ago I watched the Dallas Mavericks put research and analysis to work to win the NBA Championship. (Read this post by John Hollinger. Rick Carlisle was fantastic, and Roland Beech was the first person I ever read who put sabermetric-style analysis to a sport other than baseball.)

And I could point out all kinds of examples of Ron not using his lefty correctly, not understanding platoon splits, and not pinch hitting. I could bring up the "8th inning guy" thing again. But why be so complicated? I have an example that flies in the face of traditional baseball notions AND advanced stats, as if he was trying to be as wrong as possible.

In the 10th inning last night, he brought in Tim Dillard. Crappy, grindy fast guy Tony Campana promptly doubles to lead off the inning and is sacrificed to third by Kosuke Fukudome, bringing up Starlin Castro.

Now, in this situation, the following things are true:

1. Tony Campana is very fast. Even with an extra infielder he is pretty likely to score on a ball in play. He is the only runner that matters.

2. Tim Dillard doesn't strike a lot of people out, and induces ground balls about 46% of the time, and FBs/LDs about 54% of the time.

3. Your best chance to get out of the inning involves either a double play or a SO.

4. Starlin Castro, who cannot hit into a DP because no one is on first, only strikes out about 12% of the time. The rest of the time he either walks or puts the ball in play, and he only walks 3.5% of the time.

5. Starlin Castro is exceedingly likely to put the ball in play and drive in the winning run.

6. Aramis Ramirez is on deck. He hits GBs 35.6% of the time. This year he has struck out about 12% of the time, but for his career he has struck out 15.4% of the time.

7. Tango's win expectancy chart recommends walking the batter in this situation absent other information (i.e. who is pitching, hitting, on base, etc.)

8. You could also choose to walk Ramirez as well and pitch to Carlos Pena. This raises the possibility of walking in the winning run, but consider....

9. Carlos Pena for his career strikes out 31.3% of the time.

10. And Carlos Pena for his career, hits ground balls 37% of the time on balls in play.

In this situation, I think you should pitch to Pena. He's somewhat likely to ground into a DP, he's fairly likely to strike out, and the pitcher's spot is on deck at that point, so the Cubs would have to go to their pretty awful bench at that point. What would be even better is if the Brewers had another lefty in the pen at which point it's a no-brainer. Given that they didn't, I suppose it's defensible to pitch to Ramirez instead. It is completely indefensible to pitch to Castro.

And you know what else would have helped? Having Kam Loe available, as he truthfully has no special powers in the 8th inning, but does have a special power of inducing ground balls 56% of the time, or 12% more frequently than Dillard. But you used him in the 8th inning last night against players who are a poor matchup for his skillset, and for the 3rd night in a row, so he wasn't available.

One more fact for you. Dillard, for his career, strikes out about 5 people per nine. A strike out helps you a lot in this situation. You know who is better at striking people out? John Axford, that's who. He strikes out almost 12 per nine innings. But we had to save him for the save situation that never came to pass. At least he's well rested. (And by the way, the Cubs DID use their closer to get out of a similar jam, so don't tell me that no managers do things like that.)

I grant you that there were thousands of ways to avoid getting into this situation in the first place and I'm sure that all of these will be criticized as well, but I think this situation really shows the lack of thinking going on in the Brewer dugout.

A baseball team is a multi-million dollar investment, and wins add revenue and add value. I will never understand how owners can still entrust these enterprises to people who rely on folk-wisdom, gut feeling, and the concept of "8th inning guys".


Theron Schultz said...

I agree that pitching to Castro was a silly decision, but just a couple things to add. Pena does strike out a lot, but also walks a lot. Since he only hits RHP, I'd be uneasy with a RHP pitching to him with the bases loaded, especially if you have a walk-prone guy like Axford in there. So I would've pitched to Ramirez.

I'd be interested in seeing what effect Dillard's change in arm motion has had on his GB% or even his minor league numbers. 125 batters faced isn't that many.

Anyway, even if Loe wasn't available and Dillard is bad at inducing ground balls the Brewers have another GB pitcher: Mitre. I doubt he'd be much preferred to Dillard though.

Another concern is the pitcher's spot was up third in the next inning. The options are to burn another reliever for 2 outs and PH or double-switch. The only feasible double-switch in the 10th was Kotsay into CF, and Roenicke is willing to play him there, but I'm sure that was weighing on his mind, too. Of course, in that specific situation the OF don't really matter. I suspect Mitre would have been called in the 11th after the PH as the only remaining reliever used to going more than an inning.

It's easy to say every game should be treated like the pennant depends on it, but zooming out, the fact the Brewers don't have an off day for a while probably (rightly or wrongly) impacted the strategy, too. If they had Thursday off, Roenicke probably would have been more willing to use multiple pitchers in the 10th. Just something to think about, but how does using multiple pitchers in the 10th last night affect the next few games if Bad Narveson shows up tonight? Maybe it hurts, maybe not, but it's another thing that should be taken into account when decided whether to put in another P and who to put in.

Finally, regarding the Cubs closer being used. I'm not sure which jam you referred to, but I suspect if this game was at Miller Park, Axford would have been used and Marmol not used according to the never use your closer in a tie on the road axiom (which goes to your final point).

E.S.K. said...

Couldn't disagree with you more Theron (at least in one aspect).

Roenicke has shown a consistent and constant propensity to leave a struggling reliever in despite the outcome. I also don't honestly think Roenicke knows the schedule well enough to know when the next off day is.

PaulNoonan said...

The problem with your point about "not treating every game like it's the pennant" is that there are multiple ways to approach desperate situations, and while some leave you screwed tomorrow, not all of them do.

For instance, you do not have to pinch hit for pitchers every time. At some point you can't. But to make that decision, you have to get yourself there.

The 10th also featured the top of the order with at least Fukudome and Castro batting, and likely Ramirez. That's a good time for a good pitcher. Get through it and you can coast a bit.

You're right about Ax, and I don't necessarily like him pitching to Pena with the bases loaded, but I do like him against Ramirez, and part of the point was about pissing away Loe for no good reason. And I still like Ax pitching to Pena with the bases loaded better than Dillard pitching to Castro with a guy on third and one out. That's idiotic no matter how you proceed to handle the pen.

Theron Schultz said...

At the very least, even if you disagree Roenicke was thinking* about the ideas in my comment, they're worth taking into account when looking for a better way to manage the inning. Or I'll flatter myself by thinking that.

* - I'm neutral about Roenicke. I happen to think managers in general keep the batting order and upcoming schedule in mind during games but I could be way off.

PaulNoonan said...

I'm sure he thinks about the future, but I don't think he does a very good job of that either.

I'd be willing to defer more if not for the "8th inning guy" nonsense. If he'd properly used Braddock in the past week. If Yuni didn't play so damn much. If he got the "non-rocket science" things correct.