Culture

Lies, damn lies and analytics...

Friday, August 21, 2015Culture

A couple of weeks ago before their moment of truth arrived, the Yankees were having their merry way with the Texas Rangers leading 19-5 after six innings and the pitcher of record, either gripped by sheer boredom or in need of a beer, was allowed to depart. On came Adam Warren, one of the bright lights out of New York's suddenly revitalized farm system, to spin three fairly flawless shutout innings while his mates further padded his lead up to a reasonably safe, 21-5. Just another of those odd evenings you're going to have in a grueling 162-game -regular season, it seemed.

But what struck me most about it was the seemingly strange fact that when all was said and done Reliever Warren was credit with "a Save!"

"How?" one first asks incredulously, and then, "Why?" Was this young Master Warren's just reward for not having blown a 16-run lead? Was there something notably arduous about his task that I'd somehow failed to appreciate? Was the official scorer an old buddy from high school, or maybe his brother-in-law?

Turns out to have been none of the above but merely the absolutely correct reading of the convoluted rules that govern how a relief pitcher qualifies for a precious "Save." Clearly stated, albeit doubtless in the fine print, is the condition that if a reliever toils the final three or more innings of a game, any game, he indeed qualifies -- doggone it -- even if the score is 50-0, when he enters. So there! One quibbles with baseball's rulebook at his peril. While not quite having been etched on stone tablets it's regarded by true believers no less a body of divine dictates.

So it is that only in baseball might this discussion have merit, however remote. Only in baseball are the infinitesimal dribs and drabs of every single game -- all the thousands played every bloody season -- destined to enjoy permanent repose in the game's Doomsday Book of the infinite, arcane, and benumbing minutiae compiled all the way back to 1876 when Ulysses S. Grant was handing the Presidency over to Rutherford B. Hayes and Custer was gearing for his last stand. Talk of your history. Placing at risk baseball's statistical integrity is no idle business. Some even call it blasphemy.

Moreover, "saves," are important, even if relatively new and thereby somewhat suspect. Saves have become a major metric; increasingly seen as a measure of Hall of Fame worthiness. It may seem a bit of a stretch to suggest Adam Warren's HOF candidacy 25 or so years down the road may pivot on how many bogus saves he accrues, but the way the game is changing and the speed with which that's happening suggests someday every pitcher will be to varying degrees "a reliever." When they expand the rosters -- and that's looking inevitable -- pitching entirely by committee will be commonplace. The status of "saves" as a crucial baseball metric is about to boom still more.

Now, if statistics have always been the raw grist and sustenance of the game of baseball, their limitations have never been denied and not only because -- as in the case of "saves" -- they can be highly variable. No matter what the subject, stats can be untrustworthy because they're too easily manipulated. A sly fellow with a slippery tongue might easily draw on the same sets of stats to prove both that the world is flat and the world is round. It was in recognition of such chicanery that the venerable curmudgeon, Mark Twain, was moved to famously snort his contempt for what he called "lies, damn lies, and statistics" which he regarded essentially one and the same. If that seems harsh there's no doubt that games can be played with stats twisting the truth, and not by accident, nor do you have to be a genius to get away with it.

In baseball, it's a new breed of fanatics, raised on computers and devoted to higher mathematics, that chooses to answer the timeless suspicion of statistics with still more statistics; bigger and better ones and a whole lot more of them. Ongoing and feverish is the effort to create a body of perfect statistics that will measure and evaluate everything that can happen on a ballfield and thereby also explain why whatever happened did happen and how each player affected the outcome.

You got it? This odd new intellectual discipline, a continuing work in progress to which many smart young people are fiercely committed, has come to be known as "analytics" and to those of us who remain skeptical the adherents form what sure looks like a cult. The sheer intensity of their beliefs and passion with which they defend them tends to verify that. We think of them as being the step-sons and daughters of Bill James.

And from them we have evolving many new categories for measuring performance like, "WAR" -- for "wins above replacement" -- which (I think) basically purports to measure how much better off a team will be if player A remains healthy and keeps player B on the bench. This seems a most popular new stat with the analytics crowd and it's as casually invoked by them as "batting average" has always been leaned-on by the rest of us to indicate a player's "probable" efficacy. To this I say, "Hello!"

Another new metric one hears tossed about near as glibly is something called "DRS" which I'm told stands for "defensive runs saved". Precisely what it measures eludes me although it seems likely to have something to do with catching the ball to keep runs from scoring, the necessity of which none of us denies. And to which, I add, "Is anyone home?"

There's a long list of them -- "Whips" and "Whelps" and "Whaps" and "Whups" and "Whatevers" -- and I'm not seeking to be nasty about it because doubtless some of it makes sense and may even prove surprisingly sophisticated. But where the analytics crowd loses us traditionalists is when they totally and quite sneeringly reject the old-time metrics which have been in place since Cy Young was a rookie, as if those of us who still believe in them are totally addled Luddites. This is where and when it gets "nasty."

For example, the analytics gang abhors RBI (runs batted in). Granted some runs do have much greater value than others, and some sluggers beef up their RBI stats in one-sided routs. But overall, the producing of runs remains the central purpose of offense and thereby important and those who produce the higher numbers of them, as verified by RBI stats, tend to be the players you'd recognize as formidable even if nobody kept track of RBI.

So what's the mystery about this? It's really quite simple. You don't need to be high on logarithms to know who you would most like at the plate in the key situation of a big game when the runs that could win the game are on base begging for someone to deliver them. But a true-blue analytics devotee would reject this argument, categorically.

The quibbling bears on and it's directed at all the key statistical categories; wins and losses and earned run average for pitchers, batting average for hitters, all manner of fielding stats for defenders, even stolen bases, etc. But you get the point.

Analytics crusaders claim to be dragging the game into the 21st century where computers rule, and replays trump the ump, and soon lasers will be calling balls and strikes. But, to us non-believers it is something needless and pointless that they are doing, rather like trying to reinvent the wheel. With the end result being a sharp uptick in mere confusion.

The day will come when a chap will be canonized at Cooperstown and on his plaque will be scrolled the tribute, "He led the league in WAR 10 times".... Or ... "He set the all-time career record for DRS". Let's just say, I'm not ready for that.

- Clark Booth is a renowned Boston sports writer and broadcast journalist. He spent much of his long career at Boston’s WCVB-TV Chanel 5 as a correspondent specializing in sports, religion, politics and international affairs.