Does Size Matter (Part 5)

Over the past couple of months, we have looked in detail at how player size affects hitting performance. Today, we move on to pitchers.

In a way, pitchers are more interesting to look at than hitters. It seems intuitive that big hitters should do better than little ones. They have more power, more room to develop—that conclusion was relatively expected for me, though maybe not the magnitude. When it comes to pitchers, I have no idea.

Do bigger pitchers do better because their bodies can endure more stress? Do tall pitchers have an advantage because they’re pitching from a higher angle? Or does the advantage belong to smaller pitchers because they have more compact deliveries and explosiveness?

Today, we’re simply going to look at the variation in pitcher performance based on size, rather than the effect of size (if any) on future performance. The details of the study are as follows.

I looked at all post-World War II years. Each pitcher’s statistics (including his height and weight) were adjusted for the league average to avoid biases (pitchers are bigger today, and also pitchers allow more runs, strike out more batters, etc.). I then split the pitchers into four groups: tall pitchers, short pitchers, thin pitchers, and fat pitchers.

Pitchers were labeled as tall if they had an adjusted height of at least 6’5”, pitchers were labeled short if they had an adjusted height under 5’11”. To determine a pitcher’s expected weight, I ran a regression to predicted weight based on height. The results of the regression showed no biases against the tall or short players (that is, the residuals showed no clear pattern). Pitchers had to be at least 21 pounds overweight to be labeled as fat, and at least 21 underweight to be labeled thin. The cutoffs were chosen so that there would be 1,200 pitcher-seasons in each category. Each group has well over 300,000 plate appearances of sample size.

So first let’s simply look at performance in each group:

Tall	        4.40	6.00	2.76	0.84	0.285
Short	        4.32	5.87	2.85	0.86	0.283
Fat	        4.32	5.96	2.78	0.86	0.280
Thin	        4.53	5.51	2.69	0.86	0.282
Overall	        4.42	5.55	2.69	0.87	0.285

Lot’s of interesting stuff here. First of all, there’s not that much variance among the groups. Nevertheless, due to the sizes of the samples, the differences are for the most part statistically significant.

Short and fat pitchers seem to do better than tall or thin ones—I would guess that this is a case of selection bias. Short and fat pitchers don’t “look” good, so to make it to the major leagues they really have to be good.

We also see that three of the extreme groups are better than average, especially in the strikeouts category. Why is that? For tall pitchers, that’s to be expected, since their pitches come in from a higher angle, making for both a harder-thrown ball and for a tougher pitch to hit. Why do the thin players do so poorly with strikeouts? I would guess that they just don’t have the endurance and maybe the arm strength to power it by hitters.

Why do the tall pitchers allow fewer home runs and more hits on balls in play than any other group? This would suggest slight groundball tendencies, but a look at the 2006 batted-ball data reveals no relationship between height and groundball-rate. Either those differences are false positives, or 2006 is not historically representative in this sense. Of course, the difference is so small that even if it is statistically significant, it is not particularly significant in terms of impact. However, because tall pitchers can pitch with a greater downward slope, I would not be surprised if they did indeed tend to have slight groundball tendencies.

So how else might size affect pitchers? Do big guys have higher “survival rates” than small pitchers? There are 2,303 25-year old pitchers in my database. How do they do at 32 (an age I picked randomly)? I’ve weighted each pitcher’s numbers by the lesser number of innings pitched (or batters faced) in either season

Category	R/G	SO/G	BB/G	HR/G	Survival
Tall	        0.07	-0.30	-0.22	0.05	0.291
Short	        -0.18	-0.53	-0.25	0.03	0.292
Fat	        -0.13	-0.44	-0.21	0.06	0.345
Thin	        -0.13	0.20	0.18	0.07	0.217
Overall	        0.03	-0.48	-0.16	0.04	0.300

Survival rate is simply the percentage of 25 year-olds who were still pitching at 32. The first thing that jumps out is that the thin pitchers are completely different from the rest. Their strikeout rates actually go up, while the average pitcher loses almost half-a-strikeout a game. Their walk rates go up too. Why is that? Well, thin pitchers are much more likely to go into relief by the time they’re 32, about 50% more likely than average, probably because their bodies are more poorly equipped to handle the stress of starting in the major leagues for a long time. It’s no wonder their survival rate is low.

Speaking of which … wow. It’s clear that height has no real effect on whether or not a pitcher can shape a long career for himself, but weight certainly does. More surprisingly, fat pitchers have much more staying power than any other group. Pitching isn’t really very stressful for most body parts, except for the arm. I think that for big guys there just isn’t as much of a risk of breaking down pitching as there is hitting, running, and playing the field. Think about it: Who would you expect to fall apart first, Curt Schilling or Pedro Martinez? But that being overweight isn’t a detriment is one thing; why is it helpful?

One theory would be that some of the pitchers I bunch in as overweight aren’t really fat so much as they muscular. Maybe, but we’re looking at the fattest 15% or so; I doubt that there are enough in-shape pitchers among those to make such a large difference. Furthermore, even if we limit ourselves to the 100 most overweight players, 31 of them make it to 32. Out of the 50 fattest, 15 do. Seven out of the 20 fattest pitchers make it to 32 as well. So again, fat pitchers just have higher survival rates. Why? I don’t immediately know.

So let’s recap what we’ve learned. One, fat pitchers generally perform better than any other group and are the most likely to stick in the major leagues for a long time. Two, extreme pitchers, other than thin ones, tend to have much higher strikeout rates than average, and allow fewer runs. And three, thin pitchers are much less likely to have long careers than any other group, and those that do stay around are much more likely to end up in relief.

A Hardball Times Update
Goodbye for now.

Next time, we’ll look at how size affects a pitcher’s projection, and hopefully wrap up this series.

References & Resources
I couldn’t have done any of this without the always-fabulous Lahman Database. However, the database does not quite contain full height and weight information, so the players for whom a height or weight was not listed were removed.

Also, there was a small mistake made in calculating “survival” rates in that I didn’t remove pitchers who have yet to turn 32 from the database. Overall, this should not have an impact on the numbers, but the survival rates listed are going to be a little lower than they should be.

Comments are closed.