Not Selling Jeans or: How I Learned to Stop Worrying and Love the Walk

Whenever a Mets fan used to evoke Moneyball, it was to disparage sabermetrics. Someone would cite a statistic, someone else would say, "Moneyball ideas like that just lead to a team of fat base-cloggers! OBP is stupid!" And the former fan would dutifully retort, "You don't know anything! Moneyball was a book about market inefficiencies. Unless Sandy Alderson comes out of retirement to run the Mets, OBP is besides the point!" 

"This new SABR front office is going to turn the Mets into a bunch of stat-compiling OBP-players, who are too fat to cover Citi Field!"

"You don't know anything! Moneyball was a book about market inefficiencies. Unless--shit."

An interesting byproduct of the Alderson hiring has been a refocusing on old philosophies seemingly out of style. At his introductory press conference, confronted with a predictable line of Moneyball questions, Alderson defended his old Earl Weaver strategies. Basically, the out is the most precious capital, hitters should foremost demonstrate power and patience, and if I made it work in Oakland's Coliseum, I can damn well do it here. 

Really, much of Moneyball is about the importance of OBP, not just the movement associated with it. And in my vehement defense of the book's bigger message, I lost sight of that fact. It seems many sabermetrically-inclined fans still favor Adam Dunn-types, but only nostalgically, while believing the Nyjer Morgans of the world to be of near equal importance. 

This supposed "New Moneyball" did not replace the "Old Moneyball." Eschewing fielding for hulking on-base machines wasn't just the market inefficiency of the day. It's an organizational philosophy based on careful consideration of the alternatives, a likely directive for Sandy Alderson's Mets. 

Fans of sabermetrics are rightly eager to advance their field, but in their eagerness may have accepted some rudimentary steps as gospel, or at least, something worth building a team around. Fangraphs WAR is the ultimate number for the sabermetrically-inclined, an absolute measure of a player's offensive and defensive contributions combined. WAR, however, inherently treats one season of UZR, or an equivalent, as equal to one year of hitting contributions, which is fallacious. 

When I said I've become cynical about fielding stats, I didn't mean I'd stared at someone's UZR and said "hmm...that's too low." I've just become more realistic about their meaning. About up until James coined #6org, I accepted fielding stats on face. I've changed my mind.

My intention here is not to disparage UZR, which is still the best public fielding stat, bar none. Hopefully, if anything, I'll give skeptics ground to accept its use. I just don't think it's a "coincidence" or "bad luck" that the A's and Mariners are so bad. There are too many factors hedging against the believed importance of fielding. 

The 5% Idea

The ideas championed by Billy Beane and Paul DePodesta in Michael Lewis' book stemmed from the original Alderson front office in Oakland:

The first, short answer, according to a pamphlet commissioned by Alderson, was to spend it on hitters. The pamphlet was written by a former aero-space engineer turned baseball writer, Eric Walker. Fielding, Walker wrote, was "at most five percent of the game." The rest was pitching and offense, and while "good pitchers are usually valued properly, good batters often are not. (pp. 58)

Upon Alderson's hiring, Adam Rubin astutely called up Walker, who remained steadfast in his ideas:

You don't even have to get terribly detailed. You can see right away that offense and defense, meaning runs scored and runs allowed, have equal weight in winning ballgames. Everybody understands that. That being so, defense being partly pitching and partly fielding, it's immediately obvious that fielding is worth less than offense. It's a part of defense. If you figured that fielding is half of defense, which is wildly exaggerating it because pitching is such a big part of it, that still makes offense twice as important as fielding. That doesn't mean fielding is unimportant. But, all things considered, when you have your choices, you prefer the guy with ash rather than the guy with leather. The extent to which Sandy still believes that, I would say probably still a fair bit, but it's been so many years since I've talked to him.

The influence of pitchers on fielding is subsumed in stats like UZR in a variety of ways, such as batted-ball types and ball location, but is still largely outside the reach of public sabermetrics. Every fan knows not every groundball and not every line drive or ground ball is equal.

The Difference A Few Plays Make

Above are Orlando Hudson's fielding statistics with the Dodgers in 2009 and Twins in 2010. Hudson's a guy with a wide range of single-season UZRs (+15.9 to 2004, -10 in 2008). He's also got a reputation of being one of the game's best defenders, a consensus often challenged by some, citing said stats. 

The chart shows "balls in zone," i.e. balls that entered a portion of the field designated as his responsibility by Baseball Info Solutions, and the number of "plays," or balls in that zone Hudson played for outs. Orlando made 10 fewer plays in 2009 than 2010 in a very similar number of opportunities. He was also on pace to make about 3 more plays outside of his zone (OOZ). So in total, he made somewhere between 10-15 fewer plays in 2009. 

According to UZR, those 13 plays corresponded to about 13 fielding runs of value. In hitting terms, that's the difference between a .340 and .320 wOBA, or an .800 and .730 OPS. That's 70 points of OPS...in 13 groundballs rolling under O-Dawg's glove, up the middle. That's the difference between Jose Guillen and Justin Upton's bat in the time it took Matt Kemp to casually flip 10 balls back into the infield. Just intuitively, 13 missed groundballs is not as important as a .730 OPS hitter adding 8 homers or 30 walks to his game. No way. 

There are other inputs in UZR, but none that would explain such a marked disparity, in my mind. The 2009 Dodgers' and 2010 Twins' rotations had similar batted-ball profiles. The Dodgers had one more lefty. Between the two years, however, the real difference was Hudson. He went from converting 79.7% to 82.4% of balls in his zone into outs, which as shown, isn't a big difference. Compared to other second baseman, though, that's huge. The best second baseman last season in the AL was Mark Ellis at 85.9%, the worst, Gordon Beckham at 78.7%. 

Thirteen extra groundballs took Hudson from the basement to solidly above average. Unlike in hitting, the difference between a good and poor second basemen isn't that much, and over the course of a season, barely noticeable. I think this is why they say it takes three years of fielding data to have a sample size equivalent to one year of hitting data. 

Blissful Ignorance

Klapisch: Rough night, eh, Ike? You were 0-4 in balls in your zone.

Davis: Four balls in my zone?!? I'll show you Edina right here!

The final mitigating factor to the importance of defense is how little both we, the fans, but more importantly, the players, know about it. When a player is batting .200, he knows it. Everyone knows it. It's on the scoreboard, he gets asked about it after the game. If a player is fielding poorly, does he even know? Does he have any idea his UZR is -10, if he's got 2 errors and 20 homers? To quote a wise man: if it's so important, why don't they show it on the scoreboard?

The ultimate litmus test is Derek Jeter. Through his entire career, Jeter was terrible, while being told he was great. He posted all-time low UZRs at shortstop, but won multiple Gold Gloves. Jeter thought he was a great fielder. Then, after the 2008 season, Brian Cashman famously told Jeter what UZR and the like thought about his defense. The next year, at age 35, he posted a +6.4 UZR, by far the best of his career. The more aware players are of their deficiencies, the more likely they are to work on them. 

The Mets' Future

At the extremes, having a plus defender can be very useful. In Citi Field's expansive outfield, Angel Pagan's defense is hugely important. This fetishization of defense, however, that would put two centerfielders on either side of him who are supposedly +30 corner outfielders, is absurd. I'd put Adam Dunn and Andre Ethier on either side of Pagan and count on a combination of Pagan's range, a little luck in small sample sizes, and good positioning to mitigate against their defense on 400 flyballs, while reaping 1,200 productive at bats. You can put Nyjer Morgan and Coco Crisp out there and pray for BABIP; we'll see who wins. 

Granted, there are shades of grey between Morgan and Dunn. And granted, Morgan and Crisp are a lot cheaper than Ethier and Dunn, and for the A's, that's the point. That's a gamble they have to take. From my very Mets-oriented perspective, though, money is not the issue. So go with the proven formula. Value defense, but don't say no to Prince Fielder because he's not a good defensive first baseman. That's only 5% of what we're trying to do here.