Thursday, January 31, 2013

THE PAP MINI-FLAP

PAP is back!

What's that, you say? Isn't there enough PAP in the world as it stands? Has it ever gone away?? How can something that never went away be "back"??

Before you get hopelessly confused by this semi-philosophical sleight-of-hand, let's explain. PAP is short for "Pitcher Abuse Points." It was something of a cause celebre during the years 1998-2001, when the original crew at Baseball Prospectus developed a hypothesis about how single-game high pitch counts were destroying young pitchers' careers.

Keith Woolner: seeing red.
Rob Neyer was converted to the PAP cause in 1999; the hand-wringing about the issue increased exponentially at that point.

Skepticism about the validity of the concept in the pre-careerist world of baseball analysis caused Keith Woolner to repurpose the original system, and it became one of the calling cards that lifted him into the front office of the Cleveland Indians.

What it didn't do, however, was prove that either the approach or the data described much of what was really happening in the real world with respect to the issue.

Pitch counts for starting pitchers dropped, though no one (to our knowledge, at least) has indicated that this has occurred due to the revelations of PAP.

What's interesting, however, is that there are no studies indicating that the drop in pitches/game for starters has resulted in a material lessening of injury rates.

Bill James: greener than the Green Monster.
Bill James took that matter up in 2004, and tried at that time to show that what had come out of the pitch count analysis method was much ado about very little. (Please note that we are not saying that pitcher injuries are a minor thing, and that intelligent efforts not be made to minimize them: we are just saying that any abstract "system" to measure risk needs some more tangible basis in reality than what any of the systems or ideas about pitch counts and usage patterns have contained to date.)

But nine years have passed since the last mini-flap over PAP. Many folks have inveigled their way into baseball front offices in that time frame. There has not been much evidence that these people have revolutionized the way the game is played on the field, or that their ideas have produced any tendency for teams who utilize these analytical precepts to be more successful than teams who don't. PAP, of course, is not even part of such analytical tools--its ability to give teams an "on-field advantage" is miniscule at best. It was a beast of a different order, a kind of moral crusade with numbers.

Bill recently answered a question at his web site about pitcher workloads, and a number of folks thought that he was reinventing PAP. We looked at what he wrote and we don't really see it that way. We think that Bill was doing what he does a good bit these days at his web site--which is to talk out of his hat. (Something, in fact, that most of us do.) That's often interesting and enlightening--but it's also often a bit confounding; remember, though, that Bill is responding to questions from readers in that section of his web site, and the material appearing there is extremely variable in the amount of research/thought put into it.

Lou Whitaker (r.) to Frank Tanana: "Does anyone remember when
you owned the Big A, man? 36-18 at home from 74-77; 28-29 on the
road...and DON'T ask me how I know this!!!!"
While Bill's answer was relatively lengthy, it was not especially detailed: it fell back on the concept of extremely high single-game pitch counts (or, going further back into the sketchy past of the research conducted on this issue, to Craig Wright's primitive model of "batters faced"). The careful reader will note that Bill never suggested that a pitch count of 130, if reached a few times during a season, would be a starting pitcher's "death knell." It was the repeated application of higher-than-130 pitch games that he referenced (with Frank Tanana being the example). This is similar to the alternate theories of workload abuse that were largely ignored in the rush to judgment at the time. (During the height of the pitch count hysteria, Neyer suggested that two consecutive 127-pitch games had ruined Wade Miller's career, and that young pitchers--for example, then 20-year-old Rick Ankiel--should not be allowed to throw more than 100 pitches in a game. As we know, workload issues had nothing to do with what happened to Rick's pitching career.)

Now it's true that Bill did reference a few key elements that were prominent in the Woolner construct--particularly the "square of pitches thrown," an arbitrary construct to measure "stress" that could be replaced by several other approaches; and the notion that the most stressful pitches occur when a pitcher is tired, a conclusion that seems logical but is merely common sense.

But he also suggested another fanciful concept (not as fanciful as the type of stuff we are known to traffic in, of course; but you can't have everything). Namely: that starters should pitch every third game with pitch counts (80-90 per start) well below the levels championed by the original Prospectus crusaders.

Bottom line: this proposal by Bill is so different from what was being bandied about in those early years of hysteria that it's odd folks are trying to place it into that continuum.

Frankly, we love Bill's idea and really wish that he could convince the Red Sox management to try it. It couldn't be any worse than what's befallen their starting staff over the past two seasons. Bold, even reckless experimentation was once the hallmark of baseball: that spirit seems to have become as scarce as...well, as least as scarce as the triple.

Robert Browning: "...a man's reach should exceed his grasp,
or what's a..."--what's a beard like that going to do for you down
at the local watering hole, Bob?
So, at best, this "mini-flap" about PAP is about as "mini" as it can get. It's just part of the muddle that has befallen the field as its widespread involvement in insider baseball reaches its troubled teenage years. We don't know if Bill would really want to implement this latest pet idea of his, but we suspect that if he did, it would be well-nigh impossible for it to happen. The reach of "radical ideas" in baseball has always exceeded its grasp; despite the presence of advanced analysts in front offices, that gap doesn't appear to be getting any smaller.

[FURTHER THOUGHT: Near the conclusion of his comments, Bill makes a cogent point about what we might term the "received psychology" of starting pitchers. That psychology has not caught up with the changes in usage patterns--changes that are more encompassing than pitch counts. That helps to pinpoint where the resistance to change exists; Bill's concept was clearly designed with that in mind, as a kind of workaround for what otherwise might remain stubbornly intractable. However, in a time frame where pitching in general is gaining ground, it's unlikely that anyone will take the bait.

And then there's the fact that most reasonable analysts have learned over the course of this discussion: no single scheme can have all the answers. Pitchers (and pitchers' arms) are individualistic: it's likely that only a few of them could work in a system like Bill's. But it would be interesting to find out who could; by doing so, we might actually learn more about the true limits as they exist in the real world.]