I know that 100 isn't a completely arbitrary number, but has anyone ever done a study that makes a compelling case for a correlation between high pitch counts and injuries? I don't mean anecdotal evidence like listing a few names of guys who had a high pitch count one day, then blew out their arm a year or two later. Are there fewer arm injuries, on average, then there were 10 years ago? 20 or 30 years ago? It seems like the 100-pitch threshhold is accepted as gospel truth, but other than Pitcher Abuse Points I'm unaware of any analysis that really shows a correlation.

Baseball Prospectus’ latest incarnation of Pitcher Abuse Points, also known as PAP, tries to improve upon an initial theory that rang somewhat hollow in trying to link the number of pitches thrown with future arm trouble and ineffectiveness.

The 'somewhat hollow' links to a piece from March 2001 poking holes in the PAP theory. 2001!

100 is in fact an arbitrary number.

I just realized that Passan linked to the article that I wrote back in 2001. Wow! Never saw that coming.

I know that 100 isn't a completely arbitrary number, but has anyone ever done a study that makes a compelling case for a correlation between high pitch counts and injuries? I don't mean anecdotal evidence like listing a few names of guys who had a high pitch count one day, then blew out their arm a year or two later. Are there fewer arm injuries, on average, then there were 10 years ago? 20 or 30 years ago? It seems like the 100-pitch threshhold is accepted as gospel truth, but other than Pitcher Abuse Points I'm unaware of any analysis that really shows a correlation.

Yes.

I often get these names wrong so apologies to the authors, real and imagined.

Craig Wright did a study ages ago showing that young pitchers with lots of IP (major-league only I think) didn't age well. I think he was also the one who showed that a long string of 100+ pitch-count games increased injury risk though I've never actually read that one only seen reference to it.

BPro, around 2000-2001 or so, looked at PAP in two ways. First they looked at performance in the next start an the higher the PAP in the first start, the worse the performance in the second although this was not substantial effect. This was a trivial finding (who really cares about the impact on the next start even if it's big which it wasn't) and the one that most people picked on.

The other study was more interesting as it looked at "long-term" injury althouh, if memory serves, "long-term" here was a DL trip of one month or more. Here they, without explanation, transformed PAP into "stress" by dividing it by the total number of pitches thrown -- making it (roughly) % of "stressful" pitches thrown. This showed a reasonably strong relationship although the "statistical analysis" of it was, ummm, non-standard.

Anyway, Sean explains most of that in the article linked in the Yahoo article so just read that.

Anyway, nobody except maybe Livan Hernandez has been "abused" for the last 5 years or so. People can keep jumping up and down about BPro but this change had little to do with them. Pitch counts were adopted and enforced by old-school baseball people on a nearly universal basis. As a trend, it dates back to at least the move away from the 4-day/man rotation 30 years ago. The big recent shift was around 2000-2001 as Passan's little chart shows. But even at its recent height, there were only 212 starts of 125+ pitches in 1998 or just 7 per team.

And if anyone didn't know, PAP isn't pitches over 100. It's a non-linear function of pitches over 100 (pitches-100 cubed when I last paid attention I think). 105 pitches gives you a pretty low PAP score; 125 a very big PAP score; 140 humongous. To my knowledge, BPro has never said pitchers shouldn't be allowed to go past 100.

And, of course, every year we see a slew of sportswriters and "old school" baseball people complaining about pitch counts.

And every year we see them ignored by actual baseball GMs, managers, coaches, farm directors, minor-league managers and coaches ... every single person in every single organization in baseball has ignored this call to increase pitch count limits.

Sometime in the last couple years we even saw Leo Mazzone, a working pitching coach at the time, complain about pitch count limits ... and do absolutely nothing to change them.

I'm not saying pitch count limits are right or optimal or any such thing. I'm just somewhere between annoyed and bemused that somehow this keeps getting dumped on sabermetrics' doorstep. BPro wishes they had that kind of influence on actual MLB decision-making.

If you want a culprit, look to an insurance industry report to MLB (around 2000 I think) which claimed that pitching injuries were mostly preventable. None of the news reports I saw gave any specific info about that report so who knows what study they did or what basis they had to suggest pitching injuries could be greatly reduced.

Wainwright looked totally gassed at the end of that game, though. If injuries do actually occur more frequently when pitchers are too exhausted to maintain a clean motion, Tony should have pulled him earlier.

While I think pitchers were worked too hard in decades past, I think the pendulum has swung far too much in the other direction. There is nothing inherently dangerous about 100 pitches, or 110, or 120. The problem was that starters were too often routinely going 120 (or 130) pitches, and doing so in games (or in seasons) that were already decided. But the major problem was that young pitchers (under 24) were routinely doing this. All that was needed was a bit of attention to the area, not a sea change that led to starters being taken out after 100 pitches, and 16-men bullpens, and using 12 pitchers to get through the 7th inning and such.