Just A Bit Outside: Neyer (Quoting James) on WAR

[Orioles Jim]

In reality, he always was.

My all-time favorite quote from him:

I created a good part of the analytical paradigm that the statistical analysts advocate, and certainly I believe in that paradigm and I advocate it within the Red Sox front office. But at the same time, the real world is too complicated to be explained by that paradigm.

"It is one thing to build an analytical paradigm that leaves out leadership, hustle, focus, intensity, courage and self-confidence; it is a very, very different thing to say that leadership, hustle, courage and self-confidence do not exist or do not play a role on real-world baseball teams. The people who think that way not to be rude, but they?'re children. They may be 40-year-old children, they may be 70-year-old children, but their thinking is immature."

[DrungoHazewood]

I have to admit, like Neyer I'm not sure exactly what James is saying here. That WAR attempts to measure value, but doesn't do a good enough job? That WAR doesn't attempt to measure value? Maybe someone who subscribes to BillJamesOnline (El Gordo?) can read the whole piece that Neyer quotes and elaborate.

I've followed the discussion at billjamesonline, and I don't quite get where he's coming from here. To him it seems intuitively obvious that a halfway decent regular in almost full-time play wouldn't be as valuable as someone hitting .400 in 30 games. But it's not to me.

[DrungoHazewood]

My all-time favorite quote from him:

I created a good part of the analytical paradigm that the statistical analysts advocate, and certainly I believe in that paradigm and I advocate it within the Red Sox front office. But at the same time, the real world is too complicated to be explained by that paradigm.

"It is one thing to build an analytical paradigm that leaves out leadership, hustle, focus, intensity, courage and self-confidence; it is a very, very different thing to say that leadership, hustle, courage and self-confidence do not exist or do not play a role on real-world baseball teams. The people who think that way not to be rude, but they?'re children. They may be 40-year-old children, they may be 70-year-old children, but their thinking is immature."

I think he also knows that there are many 40- or 70-year-old children who think that the analytical paradigm is nearly useless and that all there is, is leadership, hustle, focus, intensity, courage, and self-confidence. The world before Bill James, the one without any real analytical paradigm, was full of ridiculous conclusions based on wildly inappropriate weighting of intangibles.

[Frobby]

I've followed the discussion at billjamesonline, and I don't quite get where he's coming from here. To him it seems intuitively obvious that a halfway decent regular in almost full-time play wouldn't be as valuable as someone hitting .400 in 30 games. But it's not to me.

Me neither, depending on who you are replacing the .400 hitter with when he's not able to play. If it's truly a replacement level player who is playing the other five months, then I'd think WAR is telling us you'd be neutral between six months of the decent regular and one month of the .400 hitter and five months of the replacement level guy. At least, that's what it's purporting to tell us.

[DrungoHazewood]

Me neither, depending on who you are replacing the .400 hitter with when he's not able to play. If it's truly a replacement level player who is playing the other five months, then I'd think WAR is telling us you'd be neutral between six months of the decent regular and one month of the .400 hitter and five months of the replacement level guy. At least, that's what it's purporting to tell us.

Right. The framework of WAR assumes that the guy will be replaced with... wait for it... a replacement level player. Because that's the average of all the guys who're freely available to take over when Ted Williams (or anybody) goes off to Korea. That's a reasonable framework, in my opinion. At least reasonable for making comparisons across varying contexts.

[weams]

I think James may be saying that Ted created almost all his WAR offensively over a 30 day stretch. The other fellow had a large defensive component over a season. And that there is a difference, and a difference in value, of those two individuals.

[DrungoHazewood]

I think James may be saying that Ted created almost all his WAR offensively over a 30 day stretch. The other fellow had a large defensive component over a season. And that there is a difference, and a difference in value, of those two individuals.

There is a difference in how the value was compiled. But is there really a difference in overall worth/value to their teams? Are you really better off with 30 days of nearly the best production ever and five months of scrambling to find someone to play left field, than with 130 games of solid production? I don't think it's obvious that one is more valuable than another. Was Nelson Cruz' May more valuable to the Orioles than, say, Nick Markakis' whole year?

[weams]

There is a difference in how the value was compiled. But is there really a difference in overall worth/value to their teams? Are you really better off with 30 days of nearly the best production ever and five months of scrambling to find someone to play left field, than with 130 games of solid production? I don't think it's obvious that one is more valuable than another. Was Nelson Cruz' May more valuable to the Orioles than, say, Nick Markakis' whole year?

I think Bill is indicating that the WAR overvalues the mediocre player, overvaluing his fielding and does not take into account the concentrated excellence of the 30 wonder. I do think Cruz was more valuable than Markakis last season as well. Maybe not just his May. I do also think that Bill feels that you can more easily replace a defense first slightly better than average player than create a superstar for 30 days. It's why one home run can have more value than three singles.

[Slappy]

There is a difference in how the value was compiled. But is there really a difference in overall worth/value to their teams? Are you really better off with 30 days of nearly the best production ever and five months of scrambling to find someone to play left field, than with 130 games of solid production? I don't think it's obvious that one is more valuable than another. Was Nelson Cruz' May more valuable to the Orioles than, say, Nick Markakis' whole year?

It's really an interesting question. I can totally see a scenario where one incredible month could be more valuable than a year's worth of about league average performance in the context of a season.

[Can_of_corn]

I think Bill is indicating that the WAR overvalues the mediocre player, overvaluing his fielding and does not take into account the concentrated excellence of the 30 wonder. I do think Cruz was more valuable than Markakis last season as well. Maybe not just his May. I do also think that Bill feels that you can more easily replace a defense first slightly better than average player than create a superstar for 30 days. It's why one home run can have more value than three singles.

I don't subscribe to his page but from watching him on the MLB channel I get the impression that modern advances are passing him by.

[DrungoHazewood]

I think Bill is indicating that the WAR overvalues the mediocre player, overvaluing his fielding and does not take into account the concentrated excellence of the 30 wonder. I do think Cruz was more valuable than Markakis last season as well. Maybe not just his May. I do also think that Bill feels that you can more easily replace a defense first slightly better than average player than create a superstar for 30 days. It's why one home run can have more value than three singles.

Which is odd because Bill James has said many times that most of a MLB player's value comes from being average. The above-average part of almost all players' value (if they can actually get there) is gravy. I know he's said this when discussing the weaknesses of a average-is-zero based value system.

[DrungoHazewood]

I don't subscribe to his page but from watching him on the MLB channel I get the impression that modern advances are passing him by.

There's some truth there. Win/Loss shares seems to be a lot of effort for almost no gain over what's freely available in various WAR flavors. And he sometimes seems surprised that other people have already invented things he independently figured out. But I also think he's in the position of having to reign in the excesses of the process he started. It's inevitable that some people will, for example, take the statement "we can't place a value on intangibles so we won't use them in this metric" as "intangibles don't matter or don't exist." Bill will now often talk about how there's a lot he doesn't know and won't discount, maybe even to the point of over-correcting. And now you get metrics skeptics quoting James and saying "see, even he doesn't believe this stuff!"