Having fingers, toes, and especially thumbs, has been the driving force behind our evolution. We can pick up tools and use them because we have these limbs and digits that can grab shit. So, our ancestors learned to pick up a stick, stuff it in an anthill, and catch ants more efficiently. Obviously, the next logical step was television and cell phones.
But unless you're one of those weird people, you have ten fingers and ten toes. And so we like the number ten. So that you look less stupid when you count on your fingers, we made up a number system around it. And because we like the number ten, we like the number one-hundred too. It's ten groups of ten! We express it 100.*
*When robots and computers rule the world, all pitchers will be pulled after four pitches to prevent their CPUs from overheating.
So, I will ask, in all seriousness: if we had eleven fingers instead of ten, would pitchers be conditioned to throw twenty-one more pitches per start? Is there anything that can be pointed to -- other than the roundness of the number -- to indicate that precise number of pitches to be optimal?
I omitted outings that were considerably shortened from the average and standard deviation calculation. One Andy Pettitte's injury shortened outing and one Phil Hughes terrible shortened outing were thus not considered. This was done to get a better feel of the way the pitch counts were being intentionally managed, and not "my pitcher is injured and/or mouthbreathing and simply can not be left in the game."
Due to ace status, CC Sabathia has always been given the most pitch count leeway. It's good to know that the 100 pitch concept can at least be relaxed for a pitcher as good as Sabathia, but it does seem to cast doubt on the entire process. If CC Sabathia can routinely throw 110 pitches, why must the other pitchers be limited to 100 pitches so precisely? It's also interesting to note that -- despite having the highest average number of pitches -- Sabathia has the lowest standard deviation in pitches thrown. He has thrown between 105 and 115 pitches in ten of his fifteen starts this year. He isn't being given more flexibility, his ceiling is just ten pitches higher.
So that begs the question: Would the rest of the pitchers on the Yankees staff suffer adverse effects if they were used like Sabathia? Here are some stats of opposing batters once each pitcher is beyond the 100 pitch mark:
CC Sabathia (49 PA) - .111/.184/.111 / .295 OPS /14 K / 4 BB
Phil Hughes (31 PA) - .267/.290/.400 / .690 OPS / 7 K / 1 BB
Hiroki Kuroda (19 PA) - .316/.316/.368 / .684 OPS / 4 K / 0 BB
Ivan Nova (17 PA) - .286/.412/.786 / 1.197 OPS / 3 K / 1 BB
Andy Pettitte (10 PA) - .111/.200/.111 / .311 OPS / 1 K / 1 BB
An obviously minuscule sample size, but the only Yankee pitcher who has shown signs of slowing down beyond the 100 pitch mark has been Ivan Nova. But what about a pitcher going well beyond their comfort zone? The following are the starts in which one of these five Yankees starters has gone at least one standard deviation beyond their average pitch count:
April 13th - Kuroda - 109 pitches - 8 IP - 0 ER - 5 H - 2 BB
May 2nd - Nova - 114 pitches - 6.1 IP - 5 ER - 9 H - 4 BB
May 6th - Hughes - 115 pitches - 6.2 IP - 3 ER - 6 H - 1 BB
May 10th - Sabathia - 119 pitches - 8 IP - 0 ER - 7 H - 1 BB
May 18th - Pettitte - 115 pitches - 8 IP - 0 ER - 4 H - 1 BB
May 20th - Sabathia - 121 pitches - 7 IP - 3 ER - 6 H - 5 BB
June 3rd - Hughes - 123 pitches - 9 IP - 1 ER - 4 H - 3 BB
June 7th - Sabathia - 121 pitches - 7 IP - 2 ER - 7 H - 1 BB
June 11th - Nova - 107 pitches - 7 IP - 0 ER - 5 H - 1 BB
June 13th - Kuroda - 110 pitches - 6 IP - 2 ER - 9 H - 2 BB
July 6th - Kuroda - 112 pitches - 5.2 IP - 6 ER - 10 H - 1 BB
July 8th - Nova - 111 pitches - 6 IP - 1 ER - 6 H - 2 BB
In those twelve starts, the starters combined for a 2.44 ERA and 1.20 WHIP. That's good, as it generally implies that pitchers were allowed to go deeper into games when they were pitching well. But what about any carry over effects? Here are the starts immediately following these high pitch count starts (only ten, as the last two have not happened yet)
April 18th - Kuroda - 81 pitches - 4.1 IP - 6 ER - 10 H - 0 BB
May 8th - Nova - 93 pitches - 7 IP - 2 ER - 6 H - 2 BB
May 12th - Hughes - 112 pitches - 7.2 IP - 1 ER - 6 H - 1 BB
May 15th - Sabathia - 103 pitches - 6 IP - 4 ER - 8 H - 4 BB
May 23rd - Pettitte - 98 pitches - 7 IP - 2 ER - 7 H - 1 BB
May 26th - Sabathia - 109 pitches - 7 IP - 2 ER - 7 H - 2 BB
June 9th - Hughes - 108 pitches - 6.1 IP - 2 ER - 6 H - 2 BB
June 12th - Sabathia - 109 pitches - 7 IP - 4 ER - 10 H - 2 BB
June 17th - Nova - 97 pitches - 7.2 IP - 1 ER - 7 H - 1 BB
June 19th - Kuroda - 97 pitches - 7 IP - 4 ER - 6 H - 3 BB
In these ten starts, the starters have combined for a 3.76 ERA and 1.36 WHIP. That's obviously not as good as the previous list, but that's to be expected given the selection bias. We would expect there to be a correlation between being left in longer and having a more effective start. What we are looking for here is to see if the starts that followed look like a normal selection of starts. That seems to be the case.
One interesting thing to notice: In the starts immediately following a longer than usual outing, the pitcher has reached their average pitch count total only twice (Hughes both times). Meaning that Girardi could very well be employing a "well, you threw an extra ten pitches last time out, so I'm going to look to get you out earlier tonight" kind of philosophy. Or it could be random.
On the surface, there doesn't seem to be much reason to be concerned about a pitcher going beyond 100 pitches. A pitcher being stretched out and conditioned to go around a certain number of pitches makes sense. You wouldn't want to throw David Phelps 100 pitches after a bunch of 20 pitch outings. But if you can condition David Phelps to throw 100 pitches, you can probably condition him to throw 120 pitches.
Or maybe you can't. It's nearly impossible to prove anything definitively on this subject, especially when a huge factor is the arm of the pitcher. Each pitcher has a different limit and a different comfort zone, and so a blanket statement of "starting pitchers should be allowed to pitch deeper into games" is as simplistic and incorrect as "starting pitchers should only throw 100 pitches."
I just wish that we'd use the health and appearance of the pitcher a little bit more and a round-number centered dogma a little bit less.