Originally Posted by _Sir_Charles_
Quick question Rick. WAR, in general, should it be used as a measuring stick to compare players? Ones from different leagues and/or positions? Just curious on your interpretation of how useful or important the stat is/should be.
While hardly perfect, I think it is the best tool we have for doing just that.
My problem with the dismissals of WAR are squarely on the basis that the person dismissing it never replace it with something better -- only their opaque, subjective and often arbitrary determination that leaves them impervious to criticism.
Heyman's basic philosophy appears to be this:
- MVP is a combination of production and making the playoffs
- If a player does not make the playoffs, he must be vastly superior to his competitors in terms of production
This is all well and good. I disagree on the playoffs thing, but he can use whatever criteria he wants. This leaves us, ultimately, in the same place as always -- assessing production.
At the end of the day, he thinks the best measures of production are triple crown stats and OPS. He acknowledges that Trout reaching on errors 9 times compared to Cabrera's 0 isn't meaningless, he just refuses to give it any actual consideration whatsoever. Those bases just disappear.
In terms of defense, he cites the same problem people here on RZ have: At the extremes, WAR uses defensive values that feel
too big. He doesn't actually have any specific critique of where those numbers come from and why they are probably wrong -- they just don't match what he intuits is likely accurate. Therefore... nope, no recommendation of how to objectively incorporate defense instead - just uses some kind of arbitrary fudge factor at the end. (we'll come back to this)
He says nothing about base running, other than to again assert that he doubts Trout could possibly have performed sufficiently worse as a defender and baserunner to account for the difference between last year and this year. Again, there is no incorporation of base running in terms of treating it like real value being produced and folded in to our final accounting of production.
The kick of course is the playoffs argument. Trout didn't make the playoffs, so unless he truly has been 25% more productive than Cabrera, he can't compare in terms of MVP. And because he's:
a) Discounting offensive production not captured by AVG/HR/RBI/OPS
b) Ignoring defense (at least not telling us how he counts it)
c) Ignoring base-running (at least not telling us how he counts it)
I'd LOVE to see Heyman do the following: How many runs of production do you credit each player with in terms of Hitting, Running, Fielding? Show your work. Show your assumptions. Use whatever numbers you want from whereever you want. Make them up for all I care. But show your work. Because doing this will force you to both provide numbers that can be discussed (and defended) and to show how you're layering in the value of defense and baserunning. If he wants to say Cabrera is 30 runs better as a hitter and that Trout is only 10 runs better as a fielder and 5 as a baserunner, at least we could discuss it. But the argument of "WAR can't be right because it feels wrong, Cabrera has huge triple crown stats, and Cabrera is in the playoffs" is just a plain old simple argument that does not only fails to pass any analytical threshold but in structured in such a way that it cannot be fairly reviewed and critiqued -- it's too much of a black box.
To those who claim WAR advocates such as myself treat it as unassailable, definitive proof, please cite some examples of WAR advocates taking such a hard-line approach. I'm not aware of any who do that. I think we'd do well to separate out the WAR framework from any specific WAR model. Come up with whatever values for hitting, fielding and base running you want -- but don't treat the latter two like mere fudge factors that ultimately serve as nothing more than a tie-breaker. If that is your position, if you refuse show how you account for all aspects of player production, then we don't need to keep arguing because we're talking apples and oranges.