from Amazon
Wednesday, September 26, 2007
How do pitchers age… as batters?
Someone asked me this question. He also asked me specifically about my contention about the pitchers born 1954/1955 (those pitchers affected by the DH). Here then is how pitchers (as batters) did in adjacent seasons, of pitchers born since 1955:
Age n minPA wOBA1 wOBA2 diff Chained
22 123 2496 0.149 0.178 0.029 0.109
23 222 4223 0.162 0.171 0.009 0.139
24 317 5611 0.161 0.172 0.011 0.148
25 389 7241 0.168 0.171 0.003 0.159
26 411 7667 0.171 0.168 (0.003) 0.162
27 387 7291 0.163 0.174 0.011 0.159
28 344 5956 0.173 0.162 (0.010) 0.170
29 293 5345 0.166 0.172 0.006 0.160
30 238 4825 0.171 0.175 0.004 0.166
31 216 4100 0.173 0.178 0.005 0.169
32 172 3159 0.168 0.174 0.006 0.174
33 118 2278 0.176 0.166 (0.010) 0.180
34 96 1661 0.167 0.166 (0.001) 0.170
35 71 1204 0.157 0.156 (0.002) 0.170
This is how you read the chart: at age 22, there were 123 pitchers who batted at ages 22 and 23. Take the lower of the two PA for each pitcher, and you have a total of 2496 PA for each of the matched seasons. The weighted wOBA for the first season was .149, and for the second season it was .178. That’s a diff of .029 (meaning they improved by that much from age 22 to age 23).
Let’s “force” in a wOBA of .109 (that’s part of the Chained column). All you do is take the Chained value of age 22 (.109), plus the diff value of age 22 (.029), and you get .139. So, the difference in age 23 in Chained minus the age 22 in Chained equals exactly the “diff’ column of age 22. You keep this chaining process going all the way. You see here that pitchers, as batters, peak at the age of 33!
Very old indeed. The very low number of PA makes an assertion problematic. For example, with 5000 PA, 1 SD in wOBA is .005. It would probably be fair to say that pitchers, as batters, plateau around ages 28-35. Unlike real batters, who rely on the HR for alot of their value, that’s not the case here. So, pitchers as batters probably have less volatility.
And this is for pitchers born between 1909 and 1954:
Age n minPA wOBA1 wOBA2 diff Chained
20 90 1859 0.181 0.208 0.027 0.135
21 171 4134 0.181 0.187 0.007 0.162
22 263 7328 0.184 0.184 0.000 0.169
23 406 11146 0.184 0.191 0.007 0.169
24 524 15441 0.190 0.186 (0.004) 0.176
25 621 18957 0.187 0.190 0.003 0.172
26 681 21330 0.186 0.191 0.005 0.175
27 663 21140 0.192 0.188 (0.004) 0.180
28 617 19674 0.189 0.189 (0.000) 0.176
29 544 18229 0.194 0.188 (0.005) 0.175
30 476 15819 0.192 0.194 0.002 0.170
31 409 12993 0.196 0.193 (0.003) 0.172
32 337 10822 0.196 0.193 (0.003) 0.169
33 272 8997 0.186 0.195 0.009 0.166
34 217 7152 0.195 0.186 (0.010) 0.174
35 163 5242 0.190 0.187 (0.004) 0.165
36 120 3651 0.197 0.189 (0.009) 0.161
37 88 2200 0.191 0.176 (0.015) 0.152
38 56 1605 0.176 0.184 0.009 0.137
39 39 1226 0.187 0.196 0.009 0.146
Lots more PA of course. Again though we see the same kind of plateau, though it starts earlier for these pitchers. From age 22 to age 34, pitchers as batters reach their peak.
It’s probably fair to say that pitchers today, because of the DH and therefore they have less experience before they reach the pros, do not develop their hitting “skills”. They are probably 3 to 5 years late in developing their hitting skills.
Here are the chained values for all USA born pitchers since 1955, and whether they went to college or not:
Age College nonCollege
22 0.113 0.104
23 0.133 0.139
24 0.149 0.145
25 0.166 0.153
26 0.155 0.169
27 0.146 0.176
28 0.160 0.177
29 0.157 0.164
30 0.159 0.166
31 0.166 0.169
32 0.168 0.180
33 0.180 0.169
34 0.175 0.164
I fixed the peak at .180. Sample size hurts us, but we can see that the nonCollege pitchers are a bit ahead of the college pitchers, in terms of hitting.