Bob Ryan on baseball's WAR

[3_Octave_Fart]

http://bostonglobe.com/sports/2013/03/03/bob-ryan-declares-war-war/ErbD99zVKGfn1XjonlmBBK/story.html

This is becoming as trite as the "big event host city sucks" column.
Or the "soccer sucks" column.
A terrible argument.

 

[LongTimeListener]

Which WAR is he using to say it was 9.1-6.6? B-R says it was 10.7-6.9. Of course, B-R also says Robinson Cano ranked ahead of Cabrera too, and that Alex Gordon was the 10th-best position player in all of Major League Baseball, a mere 0.7 points behind Cabrera.

But really this whole argument is incomplete for another five years until we find out what Trout's 2012 WAR was under the next adjusted formula, and then of course we have to see if it's B-R or Fangraphs that we should take as gospel truth.

 

[LongTimeListener]

Ridiculously simplistic to say the fetishization of WAR is the path to a front-office job. Whole lot more involved than that -- some very finely tuned defensive zone stats along with video scouting. Regardless, predicting what a person will do for the next five years is an entirely different exercise than evaluating what he did last year.

But ridiculously simplistic is your milieu, so I am not surprised that that would be your stance.

 

[Captain Obvious]

I get why Ryan's not into numerical modeling of players. I just don't know why people have to get so pissy when others eat it up.

 

[Dick Whitman]

The problem isn't that Bob Ryan doesn't like WAR. Plenty of people don't.

The problem is that Bob Ryan does a terrible job supporting the assertion that he doesn't like WAR.

LTL, you're too smart to reflexively support this column. You could write it 10 times better.

 

[LongTimeListener]

Uh ... I appreciate that?

I'm probably just worn out on the whole topic. I didn't think Ryan's column was bad, maybe too much on the surface. To me, a column on that topic should always include those points I (and others) have mentioned: that the two leading authorities can't even agree on what WAR is, and that it changes again and again long after the seasons are in the books. Yet even though we all know Trout's and Cabrera's WAR for 2012 will be different five years from now, we are supposed to accept them as incontrovertible conclusions in the moment.

But this ...

I get why Ryan's not into numerical modeling of players. I just don't know why people have to get so pissy when others eat it up.

... is completely opposite of how this stuff happens. I have been called a stupid old man here on this board for saying I would have voted for Cabrera. There were dozens, maybe even hundreds, of columns deriding the 22 Flat Earthers who wasted their vote last fall. Probably the most egregious was Passan's saying that because Nate Silver nailed the presidential election, Cabrera voters are wrong.

 

[RickStain]

I'm ready to just give up on WAR so I don't have to deal with these absurdly bad columns. It's a lazy, shorthand stat that no one really needs, although it is convenient. It's the fast food of analysis. I like it, but I can certainly do without it.

There's no war inside baseball. The war is entirely being fought by fans (and that includes writers).

 

[Knighthawk]

WAR as a concept is fine. It's a very useful way to measure players in a context-free way. I don't think the MVP should be decided in a context-free manner, but that's just an opinion.

The elephant in the room is that a big chunk of WAR is junk, because no one has found an accurate way of measuring defensive value. Fangraphs thinks that Jhonny Peralta was one of the best defensive shortstops in baseball last season.

 

[Bubbler]

 

[RickStain]

It's not that we don't know how to quantify defensive value. It's that you have to make choices no one likes to make about what you are trying to measure. The best defender isn't always going to be the most valuable.

 

[93Devil]

OPS is a perfectly fine way to judge a hitter.

Fielding is trickier, but what about making a rating that measure how many times a game a player catches a batted ball? If a player does have limited range, it will show up in that stat. Compare shortstops to shortstops or RF to RF and the ones with good range should start to rise to the top over the course of a season and a career.

And errors are errors.

 

[Michael_ Gee]

Fielding is the Lost Dutchman gold mine of sabermetrics. I'm of the opinion that a consensus on the most accurate means of measuring fielding ability will NEVER be found, if only because the people inside baseball, managers, scouts, front offices, etc., place a much higher value on consistency than they do on range, while outsiders do the opposite.

 

[FileNotFound]

I've always kind of thought "range" = "how well a player recovers from being out of position against a given batter."

Bad fielding is pretty easy to measure. Determining who is better between two good defensive players really does rely as much on an eyeball test as measureables.

 

[Dick Whitman]

OPS is a perfectly fine way to judge a hitter.

Fielding is trickier, but what about making a rating that measure how many times a game a player catches a batted ball? If a player does have limited range, it will show up in that stat. Compare shortstops to shortstops or RF to RF and the ones with good range should start to rise to the top over the course of a season and a career.

That's a remarkable idea. I wonder if anyone has ever thought of it before now. Make sure to alert the editors of the forthcoming 1969 Baseball Encyclopedia.

 

[RickStain]

Fielding is the Lost Dutchman gold mine of sabermetrics. I'm of the opinion that a consensus on the most accurate means of measuring fielding ability will NEVER be found, if only because the people inside baseball, managers, scouts, front offices, etc., place a much higher value on consistency than they do on range, while outsiders do the opposite.

The problem (well, one of the biggest problems, anyway) is that chances are so variable from player to player for each season. No one wants to here that some mediocre fielder had the most valuable fielding season because a ton of balls were hit his way and he fielded them, or that some super-elite gold glover wasn't that valuable defensively because few of the balls hit in his direction were difficult chances this year.

 

[Versatile]

OPS is a perfectly fine way to judge a hitter.

So in 2000, Jeffrey Hammonds was a better hitter than Derek Jeter, Magglio Ordonez, Roberto Alomar, Scott Rolen, Andruw Jones, Andres Galarraga and Phil Nevin?

Fielding is trickier, but what about making a rating that measure how many times a game a player catches a batted ball? If a player does have limited range, it will show up in that stat. Compare shortstops to shortstops or RF to RF and the ones with good range should start to rise to the top over the course of a season and a career.

That stat already exists. It's called total chances, and it's largely worthless because teams and fields aren't all the same. Playing left field in Fenway Park is very different than playing left field in Petco Park.

And errors are errors.

Errors are what a given scorekeeper on a given day decides are errors. The variance in what gets called an error has been proven again and again.

 

[Dick Whitman]

One of my favorite new stats is this Win Value Added that got bandied around a little bit last year during the Trout-Cabrera brou-ha-ha. While WAR is context-neutral, which bugs a lot of people - justifiably - WVA is completely context-based. Essentially, every play in a baseball game adds or subtracts each team's chances of winning that game by a certain percentage. And a player's WVA over the course of a season and it tells you what value he added to his team, for real, rather than hypothetically like WAR.

Way better than WAR, to me, for determining an individual award in a season. Because I think that What Happened matters, i.e. hits at key times and such.

 

[SoCalScribe]

The problem with WAR is that it is based on a nebulous and poorly defined concept that varies from year to year. Same issue with VORP, obviously.

Why they can't use the same exact idea and just peg it to a median or mean, or even a certain percentage thereof, I do not understand.

A concept can't get much more nebulous than "replacement" since it is inherently obvious that "replacement" levels are so massively different between the best teams and the worst teams. To imply otherwise for the sake of a statistical model is asinine, and therefore, they should use deciles or other tiers rather than defining a "replacement level" that can never actually be approximate to more than a handful of teams' minor league talent levels in a given year.

 

[Dick Whitman]

A concept can't get much more nebulous than "replacement" since it is inherently obvious that "replacement" levels are so massively different between the best teams and the worst teams. To imply otherwise for the sake of a statistical model is asinine, and therefore, they should use deciles or other tiers rather than defining a "replacement level" that can never actually be approximate to more than a handful of teams' minor league talent levels in a given year.

People, including Ryan, get so worked up about the concept of "replacement." I find it the least objectionable thing about WAR. It requires a base line. That's all. Objecting to it is like objecting to the existence of the number zero. If you made replacement values team-specific, then the players wouldn't be measured on an equal footing against one another. Which defeats the purpose.

 

[Azrael]

Posnanski.

"Zaidi and I were talking about this when he told me something that I found utterly staggering. He said that Oakland’s objective model for measuring a player’s value — remember now, we are talking about the Oakland A’s, the Moneyball people, Jonah Hill and so on — found that Miguel Cabrera, NOT Mike Trout, was more valuable in 2012.

Well, that’s not exactly right. He was quick to say that the difference between the two was so slight as to be almost invisible — they were, for an intents and purposes, in a virtual tie. But their system did have Cabrera ahead by the tiniest of margins."

hardballtalk.nbcsports.com/2013/03/04/revisiting-trout-vs-cabrera-mvp-debate-with-a-twist/related/