The Boston Globe

Sports

Bob Ryan

WAR stat in baseball is complete nonsense

Was Mike Trout the best all-around player in baseball last year?

You kidding? Who doesn’t know that? No one combined offensive and defensive skill to the extent he did. It’s barely a debatable topic.

Comments

I knew baseball stats were often useless when I was ten. The ball drops in front of you because you weren't paying attention? Not and error?

Wait.  You hate WAR, but it supports your argument that Trout is better than Cabrera?  What?

There's no war going on in baseball. The war is over. In front offices all across the American and National leagues, analytics won. The only war taking place is in the minds of the fans, and the media, and no tired clichés about mother's basements are going to change that.

Yeah, closing your argument citing Trout's superior WAR is a little odd. But then again, so was your support for the lie otherwise known as "The Patriots filming the Rams' Walkthrough." Yep, we remember that.

I agree with the point that Bob Ryan is trying to make, but I don't believe that he has provided a good example.  He is agreeing that WAR provides a good statistic for evaluating who the best player is, but that player might not be the most valuable to his team.  That makes perfect sense in baseball or other sports.  For example, a good defensive outfielder is much more important if he is on a team that has mostly flyball pitchers.  When we talk about the contribution that a player makes to a particular team, we need to consider the roles that the player has on that team, and how much better he is than the player who he would actually replace.  Of course, we also have to consider that the team would have a chance to pick-up an average replacement player if the player in question was hurt or left the team.

 

I don't agree with the example of Cabrera and Trout, however.  Trout is so much better than Cabrera that I do not believe that Cabrera is more valuable. 

 

I'll give Ryan credit for trying to keep up with baseball statistics.  He does a much better job than most of the writers in his generation on this.  However, when he comments on who is a good player and who is not, he he often reverts back to old-fashioned statistics that are not useful. 

Good of Bob to make his comparison of the two players while totally disregarding the fact that, by any measure, including untrained eyeballs, Trout was a near spectacular defender and Cabrera may have been the single worst defender at his position in MLB.  That, of course, is such an inconvenient truth that Bob felt compelled to ignore it.  

WAR, is not a perfect way to evaluate baseball accomplishment nor is it necessarily predictive of future outcomes, but it is a well thought out metric designed to give us deeper insight into a game I have loved for most of my 68 years.  How can that be a bad thing?  Bob's sad attempt to obfuscate how WAR is constructed does us no service and is unworthy of the best sportswriter this town has ever known.

Rather than compare a player to a hypothetical construct (a replacement-level player), I'd rather see the comparison with the average player at a given position.  Superior players should have very high WAA (wins above average) and it would be a more informative number than wins above a fringe player. 

Replies

It's been tried, but major league average is actually pretty exceptional.

Ryan is right, the WAR stat is a joke. It simply does not measure opportunity and environment, i.e. the same argument that people use to bash RBI can be used to tear down WAR. It often measures context more than performance and not a single person who casually refers to WAR when discussing a player's value can actually explain what the hell it is.

As for Trout not winnign the MVP, yes, he should have won, but his snub is far less of a travesty than Elsbury losing to Verlander. Heck, Elsbury's numbers, side by side next to Trout, are almost indentical and he lost to a pitcher, not a triple crown winner.

Replies

Ellsbury's numbers in 2011 are much different than Trout's in 2012. Trout is much better for hitting (OPS+ of 171 vs. 146), fielding (defensive runs saved per 1,200 innings above the average player of 31 vs. 6 for Ellsbury) and base stealing (Trout stole 49 and was caught 5 times, Ellsbury stole 39 and was caught 15 times; regression analysis shows that break-even is about 2 steals for each caught stealing). Given that Trout was much better for hitting, fielding, and base-stealing, what do you believe that Ellsbury did for you to believe that thier numbers are "almost identical"?

bob--you've been my favorite writer for years and i sit almost right next to you in sect 18.  never thought i'd say this about you,but you sound like a neanderthal republican right-winger: denying all facts and objective knowledge for no good reason at all;just because you feel like it? please dont become a baseball birther and climate change denier. if you do,my faith in sports journalism will be altered forever

WAR is a construct, a framework. Different inputs can be used, for different purposes. If you want to argue a particular result, the issue usually lies with a particular input, which in the case of WAR is usually defense.

 

Attacking WAR on the basis of the concept of replacement player shows that Ryan very fundamentally does not know what he's talking about. You can peg the comparison level anywhere you want -- replacement player, average player, top player -- and it won't make one bit of difference in the result.

 

In this column Ryan is screaming "get off my lawn", but can't even really say why. Ultimately, this is just an incoherent complaint that there's something he doesn't understand, and he doesn't like it.