Wednesday, November 16, 2011

THE AWARD YOU DON'T WANT TO WIN

A socio-linguist studying the sabe-centric world might well have a tough time picking out the most prevalent (read: privileged) jargon--what H. L. Mencken (channeling his inner Victor Hugo via Thorstein Veblen) would have dubbed "the argot of the analyst class."

"Sweet Jehosaphat, Ann, every week our relationship regresses
to the mean!!"
There are so many to choose from...

--But let's not dwell upon this, as such a discussion simply revisits the peculiar intractability inherent in the ongoing war over WAR, the guerrilla infighting, the race to phantom regions of moral rectitude, the shameless borrowing of social science concepts for the sake of intellectual carpet-bombing, etc.

Whatever lists of involuted supercalifragilistic expialadociousness are compiled to tourniquet the machinery of the "meme" as it has spread across the little world of baseball analysis during the past three decades, there is one phrase that's almost certain to be at the top. What is it? No, it's not that girl, it's...

Regression to the mean.

Fuggedabout that girl...this girl has got the stuff. While her affinity
for upscale malt liquor is elevated, there's no truth to the rumor that
she'll be naming her new band LoLo and the Suds...but, hey, if
Lauren Hillman keeps writing songs like "Young Love," she can
do whatever she wants....
This is the allspice of sabermetrics, even seems to not eff up the taste of ice cream when you accidentally take off the top and dump instead of sprinkle. Of course, tastes differ, but this explanation of events seems so ladled with preservatives that you can not only leave it on the bedpost overnight, but you can leave it out for decades...aeons...and it won't spoil or--worst of all--get clumpy due to prolonged exposure to the air.

And that's just what we need in a phrase that combines precision and puffery, rigor and mortis, warp and woof.

Now it turns out that there is one post-season award that exemplifies the actual principle within the phrase "regression to the mean." It's an award whose trophy should contain--or possibly simply just be--a double-edged sword.

"Managers of the Year" by Francisco Goya...
What's that award?

It's called Manager of the Year. Today, two fine fellows will get honored for their work in the dugout. Next year, they will almost certainly get buried.

Think I'm off my rocker? (It's OK, national polls favor your position.) Here are the facts: the winning percentage of managers in the years they win the MoYA (rhymes with Goya, so...) is .591. Their winning percentage in the year after is...

.511.

That's an eighty-point drop. Teams win 13.4% fewer games in the season following a year where the manager has been a MoYA.

There have been fifty-four managers who were MoYA and managed again in the next year. (We tossed out Bobby Cox, TOR, 1985, and Davey Johnson, BAL, 1997, because they didn't manage in the following season). Out of this group of fifty-four managers, only four of them (7%) had a better winning percentage in the year after they won the MoYA. Those four managers who've beaten the odds are: Jim Leyland, PIT, 1990; Bobby Cox, ATL, 1991; Gene Lamont, CHW, 1993; and Joe Torre, NYY, 1996.

Everyone else on the list (and it's reproduced for you at left) has, to some degree or another, taken it in the tukus. As is usually the case, Tony LaRussa is prominent on this list, and just might walk off with the record for the longest span of MoYA, having gotten his first in 1983 (and winning 25 fewer games the next year). We all know that Tony is singular, and so it shouldn't be surprising that he is the only manager to win the World Series in the season following a MoYA: 1989. The A's did win fewer games in '89 than in '88, but Tony is, as always, at least a partial anomaly unto himself.

This partially explains why it is so rare for anyone to be MoYA in successive seasons. Over the past twenty-nine years, this has happened only once, when Bobby Cox won in both 2004 and 2005.

Who had the greatest percentage drop from one year to the next? Until the conclusion of the present season, the cruel fact was that the MoYA who crashed hardest was--you guessed it--someone who worked for Kansas City. Tony Pena, who also holds the record for winning a MoYA with the lowest seasonal WPCT (.512, 83-79), watched in Goya-esque horror as his team tumbled into Boschian regions (Don or Hieronymous, take your pick...) the next year, going 58-104. That amounted to a 30.12% drop in WPCT.

As we said, until this year. Ron Gardenhire has taken Pena off the hook with an even more fearsomely prodigious swan dive, one that represents a drop of 32.98%.

Tony LaRussa managed to have two follow-on MoYA disaster years--1984, noted already (a 25% drop) and 1993 (a 29% drop). One suspects that Tony knows all this, and here's yet another reason for him to ride off into the sunset in case the BBWAA hands him another double-edged sword.

Looking from the insider perspective, the one where the folks involved actually put on the jockstrap, there's a phrase that resonates with the concept of "regression to the mean." That phrase, uttered on behalf of Sandy Koufax by the great sportswriter Ed Linn (the ghostwriter of Sandy's 1966 autobiography), goes as follows:

"This game can't wait to humble you."

That apparently applies to managers even more than players.

So congrats to the 2011 MoYA winners [UPDATE: Kirk Gibson and Jumpin' Joe Maddon]--and...

Condolences.