Daily Box Score 8/26: Counting Pitches

Ah, pitch counts: the one subject that threatens to cleave the sabermetrics community in two. Proponents of strict pitch counts get to claim they are protecting the health and careers of young pitchers. Opponents of pitch counts argue that they've ruined the game and counterintuitively made generations of young pitchers more prone to injury. 

I'm not saying it isn't a good argument to have, but it's not something I've got the patience or courage for today. I'd rather look at some of the less obvious aspects of counting pitches.

Table of Contents

Effectiveness by Pitch Count
Knocking a Pitcher Out of a Game
The Subjective Experience
Discussion Question of the Day

Effectiveness by Pitch Count

Let's start by asking a question. During which bucket of pitches, in increments of 25, is a starting pitcher most effective? I'll give you a second to think about it.

Your reasoning probably went something like this (if I may invade your mental events for a moment):

1. Pitchers have the least fatigue at the beginning of the game. Maybe it's pitches 1-25?
2. No, that's ridiculous. Too many runs score in the first inning for that to be true. Besides, maybe some fatigue is good.
3. Pitchers settle into a groove, and that groove usually starts around the third inning.
4. By the sixth or seventh inning, the pitcher is usually starting to experience fatigue.
5. So, let's say innings three through six, that would either be pitches 26-50 or 51-75.
6. What the heck? Let's say 26-50.

And then maybe some of you cheated. It turns out, in the 2008 American League, pitchers were most effective throwing pitches 26-50. In fact, though, in 2009, 2007, 2006, and 2005 (when I got tired of looking), pitchers OPS against was as low or lower over the bucket 1-25 as it was for 26-50. So, that's odd, but perhaps it can be explained by the presence of relievers, who in the aggregate have an advantage over starters in terms of raw numbers and who rarely throw more than 25 pitches.

Then I saw this neat little post from the other Tommy, over at DRaysBay. He's writing about a comment by Rays' former Mets' pitching coach Rick Peterson, who made the argument that pitchers begin to decline rapidly in performance after the 100th pitch. And while that's an interesting discussion, it veers too close to the sabermetric cleavage I am trying to avoid. 

But as I was reading the post and looking through the numbers, something struck me about the Rays starters. Nearly all of them performed best during the bucket of pitches that stretches from 51-75. Shields, Garza and Sonnanstine are clearly best from 51-75. Niemann is slightly better from 26-50, and Price more dramatically so. But it's within shouting distance. 

This, I think, is interesting, although I'm not sure quite what to make of it. Of the 2009 AL ERA leaders, Greinke does not exhibit the trait (the guy is best from 76-100, I mean who does that?), although Felix Hernandez, Edwin Jackson, and Roy Halladay do. Perhaps it is just the case that the best pitchers are the ones who can perform best against a lineup the third time through, rather than the first.

Is this result intuitive? Is it just complete randomness? What do you think?

Knocking a Pitcher Out of the Game

Conventional wisdom says patient teams do better because they wear the starting pitcher down more quickly, causing the opposing manager to go to his bullpen more quickly. But MGL wonders if this is necessarily a good thing:

What if you have an average or worse starter on the mound?  And roughly half of all starters fit that bill.  With what we know about how much poorer pitchers do as they cycle through the order and with what we know about how much better relievers are than starters (at least partially because they usually don’t face the lineup more than once), maybe getting the starter out of the game early is NOT a good thing for the other team.  Not to mention the fact that once the relievers start coming in, the pitching team’s manager can attempt to get the platoon as much as possible.

I think it's important to make clear what we're talking about. In one sense, MGL is absolutely right. If the other team started a bad pitcher, you don't necessarily want him out of the game. But managers tend to remove pitchers based on variables that include, but are not limited to, pitch counts.

For example, it is very rare for a manager to remove a pitcher unless he is struggling or at or above 100 pitches. In that case, it is almost always a good thing when a pitcher is removed from a game before 90 or so pitches, although this has nothing to do with being patient. Rather, it's simply a good indicator that the starter stunk up the joint.

The Subjective Experience

Unless you have a mammoth collection of RSS feeds about baseball, or are a Phillies fan, you probably haven't been following Phillies pitching prospect Michael Schwimer's diary over at Phuture Phillies. Never fear! It just so happens that I do, and I am, so I have been. He's got a bit of a sabermetric bent, he's very self-reflective, and he writes clearly. 

But yesterday's entry was undoubtedly his best. He broke down a two-inning appearance he made on August 20th, 2009. He was awarded the save and recorded five strikeouts while allowing two hits, a walk, and zero runs. But it's so much more interesting to hear his pitch-by-pitch account of the appearance. Here's an excerpt:

PITCH 17: I am not going to waste any time, and I go straight for the kill. He will definitely swing at a good up and down slider that is a strike for 50 feet then drops. I execute the exact pitch I want but he spits on it (Spits on it is a term that describes how a hitter take a pitch. To spit on a pitch means the hitter did not even attempt to offer at it. It is as if he has the take sign). Now I am a little flustered. I got the opposite reaction I was looking for. The only possible reason for that is either he has an unbelievable eye or he was in take mode. You might say to yourself, why someone would be in take mode with the count 0-2. My answer at the time was that he knew I was trying to get him to swing and miss, so he is trying to bait me into throwing the next pitch a little higher, and right in his wheel house. Count 1-2

He also includes links to the innings he pitched, so you can follow along with him as he describes his thinking for each pitch. I highly recommend the entire diary; it's a rare and wonderful peek into the mind of a pitcher. He even has a grand theory of pitching:

As you can see, I treat pitching as much a mental exercise as a physical exercise. The key to pitching is to connect the two. Throughout this outing I was wrong a few times with regard to what pitch should be thrown to which location, and I miss executed some pitches. The majority of the time I was correct in my thought process and was able to execute the pitch properly. My goal is to be able to do this all the time.

I gotta say, that seems like the right goal. 

Discussion Question of the Day

For a little fun, here's a great list of the ten best baseball audio moments. Bugs Bunny, George Carlin, and James Earl Jones all make appearances.

What aspect of pitching most puzzles you? When do you most want to get inside a pitcher's head? 

I always wonder whether pitchers are furious when their manager instructs them to intentionally walk a batter.