The problem with WAR
By Jesse Pantuosco, Fantasy Sports Writer
Philadelphia, PA (Sports Network) - War, what is it good for?
We're about to find that out.
And no, I'm not referring to Barack Obama's handling of foreign policy. This is about W-A-R, a sabremetric I assume was created by Bill James and his team of evil henchmen in a laboratory somewhere deep beneath Fenway Park.
If you didn't already know, WAR stands for wins above replacement. Basically it's the baseball equivalent of PER, the player efficiency rating pioneered by NBA talent analyst and current Memphis Grizzlies vice president of basketball operations John Hollinger.
Baseball's sabermetrics trend has gone mainstream thanks to Michael Lewis' best-selling book "Moneyball" that came out in 2003 (an Oscar-nominated film by the same name hit theaters eight years later). So mainstream, in fact, that sites like ESPN and Fangraphs are beginning to devote entire pages to it.
Want to know Miguel Cabrera's isolated power average? We can calculate it. Curious as to how many runs Curtis Granderson generates every 27 outs? We've got that too.
Nothing goes unmeasured in the world of sabermetrics. The amount of stats we have today is practically endless. And we can use this data to paint just about any picture we want.
Sabermetrics in its simplest form can be boiled down into one key measurement: wins above replacement. This all-encompassing collection of data assigns an overall win value to a given player, usually ranging from zero to six. Players who are especially good can sometimes crack double-digits (Mike Trout did this in 2012) while players who perform below our expectations can sometimes dip into the negatives (Kansas City's Jeff Francoeur was an embarrassing - 2.7 last year).
Because wins above replacement takes so many different categories into account (it combines hitting, base running and fielding statistics), it's probably the most complete statistic we have to measure a player's overall worth.
But is it something that fantasy owners should really be paying attention to?
Well, take a look for yourself. Here are ESPN's top ten players in WAR from last season.
1. Mike Trout: 10.7
2. Robinson Cano: 8.2
3. Buster Posey: 8.2
4. Andrew McCutchen: 7.0
5. Miguel Cabrera: 6.9
6. Ryan Braun: 6.8
7. Yadier Molina: 6.7
8. David Wright: 6.7
9. Adrian Beltre: 6.6
10. Alex Gordon: 6.2
It's a pretty interesting group, isn't it? Cabrera and Posey were MVPs and Trout probably would have been as well if he hadn't started the season in Triple-A. Cano, McCutchen and Braun, though not in the MVP discussion are all likely to be drafted in the first round of fantasy leagues later this spring.
But then we have two wild cards: Gordon (.294, 14 HR, 72 RBI) and Wright (.306, 21, 93). Both players would fall under the "good but not elite" distinction. Taking either of these players in the first two rounds of a fantasy draft would seem like a stretch.
Things get even fishier when you consider that ESPN and Fangraphs don't even measure WAR the same way. Look at Fangraphs' top ten from 2012:
1. Trout: 10.0
2. Posey: 8.0
3. Braun: 7.9
4. Cano: 7.8
5. Wright: 7.8
6. Chase Headley: 7.5
7. McCutchen: 7.4
8. Cabrera: 7.1
9. Jason Heyward: 6.6
10. Beltre: 6.5
If there's no universal measurement in place, how can we distinguish between ESPN and Fangraphs' rankings? It's impossible.
Here is where things get even weirder. Albert Pujols and Josh Hamilton are nowhere to be found in either top ten.
Pujols (.285, 30 HR, 105 RBI) is ranked 44th in WAR on Fangraphs (3.9) behind players like Nick Swisher (.272, 24, 93), David Murphy (.304, 15, 61) and even Baltimore's Matt Wieters (.249, 23, 83). Josh Hamilton (.285, 43, 128), ranked 34th on Fangraphs with a WAR of 4.4, is slotted behind Angel Pagan (.288, 8, 56), Miguel Montero (.286, 15, 88) and Bryce Harper (.270, 22, 59).
According to Fangraphs, players with a WAR between 0-1 are considered "scrubs," players in the 1-2 range are "role players," 2-3 represents a "solid starter," 3-4 is "good," 4-5 would be an All-Star, 5-6 is reserved for "superstars" and six and above would be MVP level. By this estimation, Mark Teixeira (.251, 24, 84) and Carlos Gonzalez (.303, 22, 85) are merely "solid starters" while Austin Jackson (.300, 16, 66) is a "superstar."
It makes sense why a player like Prince Fielder isn't in the top ten. Though he's an excellent hitter (.313, 30, 108 last season) he's always been a pitiful defender (11 errors in 2012, most among AL first baseman) and he's produced just three stolen bases combined over his last three seasons. Because of these deficiencies, Fielder has only produced a WAR higher than 4.5 once in his career.
But Teixeira (79th on Fangraphs) and Gonzo (83rd) are both Gold Glovers. How could the WAR formula overlook something that?
Maybe it was naive of me to assume that WAR would be my cheat sheet on draft day but the numbers just don't add up.
I don't completely discount WAR and overall, I'm a fan of the sabermetrics movement. Billy Beane and Bill James gave the sport an added dimension by uncovering all of these new statistics.
But am I crazy enough to draft Ben Zobrist ahead of Matt Kemp simply because Zobrist has a higher WAR? Not a chance.
02/07 14:41:11 ET