header image

header image

Monday, January 13, 2014

The Cold Hard Stats


In the weeks leading up to the annual tallying of Hall of Fame votes, two topics of conversation begin to intensify.  One is whether or not a known, or suspected, user of performance enhancing drugs is worthy of selection.  If you're like me, you've heard enough about that topic to last you for a while.  The other conversation is a recalling of statistics.  Reporters are quick to tell us how many 200-strike out seasons a pitcher accumulated, how many times a hitter led the league in batting average, and so on.  Those who put in a little more research, will provide insightful context such as "in a decade dominated by hitting, Pitcher X struck out 200+ batters seven times" or "Player Z's batting average was 30 points higher than the league average for five years in a row".   This listing of stats, and the accompanying analyses, is useful. It allows us to compare how players performed in relation to their contemporaries, and, to an extent, in relation to players of different eras.  But we have to be careful with stats.  They don't tell the full story.


One of my favorite parts of writing this blog is the response I get from readers.  For example, after I posted the article about Roberto Clemente, I got multiple replies from a slightly older generation of fans who remember watching Clemente play.  They recalled his fluid throwing motion, his confident swagger, his willingness to take chances - and deliver - on risky plays, and his ability to generate excitement.  No one mentioned his batting average or on base percentage.  His stats, impressive as they were, were not what made him great.  To understand that, you had to watch him play.

As a kid, I greatly admired the playing style of stolen base king Ricky Henderson. When Rickey got on first base, competent pitchers fell to pieces.  They were so concerned that he could break for second base at any moment, they weren't thinking about the batter.  They were looking over to first and sizing up his distance from the bag. When they finally did pitch, they often rushed their delivery.  Rickey had them right where he wanted them and the next batter, plus his entire team, benefited.  Just by drawing a walk, Rickey Henderson could change the entire game.  What stat measures that?

In recent years we've been introduced to a slew of new advanced "stats" (also called Advanced Metrics) that are designed to give us a more objective way of looking at, and comparing, players.  A pitcher with a great defense behind him will give up less runs and earn more wins than a pitcher left to the mercy of a poor defense, so ERA and Wins by themselves are not a completely reliable measure of a pitcher's true ability. Similarly, the batter who hit directly behind Rickey had more opportunities for RBIs than other, less fortunate, hitters.  Advanced stats are meant to even the playing field.  They attempt to make adjustments for ball park dimensions, minimize the distortion of a team's defense on pitching stats, and even try to quantify a player's "luck" (a ground ball just out of an infielder's reach is a "lucky" hit, while a ground ball hit right at a defender is "unlucky").  Sports writers like to break out these stats at the end of the season to justify, or criticize, the awarding of Gold Gloves, MVPs, etc.

One of the more popular advanced stats right now is WAR (Wins Above Replacement).  The intent of this new metric is to assess how valuable a player is over (or under) the value of the average replacement player.  It takes into account multiple aspects of the game (hitting, fielding, base running, defensive position, "luck", etc.) in order to produce a number that can be compared to any other player at any position from any era.

You can make up your own mind about the validity of  advanced stats, but there are a few things you should keep in mind.  One is that many of them don't actually measure anything.  They are hypotheses.  We have no way of knowing how a pitcher's ERA would have looked with a different defense behind him.  And doesn't a good pitcher make adjustments to external factors?  As for WAR, it is a hypothetical stat that is a composite of several more-specific hypothetical stats plus other subjective variables, making it a hypothesis based on hypotheses.  Furthermore, there is no agreed-upon formula for WAR. Most stat-tracking websites (Baseball-Reference, Fangraphs, etc) have their own formula.  When you look at the factors for calculating these un-quantifiable qualities, what begins as an attempt to compare players more objectively starts to look extremely subjective.

Whenever a sports writer claims that Center Fielder X shouldn't have won a Gold Glove because some of his league counterparts had better advanced fielding metrics, my first reaction is to wonder whether that writer has ever actually watched a baseball game.  Maybe the writer ends up being correct, but it's just as likely that the voters were taking into account intangibles not measured by stats... the fielder's natural instincts, his ability to read hitters, the way he leads his teammates on the field, and/or other factors that can only be determined by observing the player.

Therefore, I'd like to propose a new stat.  I call it PEACE, which stands for Player Evaluation Actually Captured by Eyesight.  It's not a stat for players.  It measures the ability of fans, sports writers, HOF voters, and others to determine the true worth of a player.

If you really want to evaluate players and enjoy the game, WAR is not the answer.  Give PEACE a chance.  




Photos: 1. the back of a Greg Maddux card, 2. me standing in front of a Roberto Clemente exhibit at the National Baseball Hall of Fame and Museum, 3. a collection of Rickey Henderson memorabilia at the National Baseball Hall of Fame and Museum, 4 & 5. homemade creations using Microsoft Paint and free clip art.