Is the WAR war over?

[jojo]

No.

Why don't you start by actually providing specific reasons that support the several strongly declaritive statements you've made in this thread?

We've had multiple conversations on this issue over the years so I'm confident you can find the reasons for my argument among the thousands of words I've authored in the archives on this subject out of respect for those who who either questioned the argument or were curious about the topic.

I'm still struggling to find the reasons behind your strong opinions on this issue. Please point us to them.

[edabbs44]

Why isn't it? We don't measure hitting ability based on the "chances" they get based on the ability of the pitchers they face throughout the season (though there have actually been things done on that premise and some guys very clearly have advantages in the competition they face versus others).

It seems that you want to equate UZR with skill level, but we don't do that same thing with hitting.

No, but I would be really interested in hearing about stuff like that since I have posted a decent amount about that topic when it came to one guy in particular.

But I feel like UZR is similar to using counting stats when comparing two players with different ABs, different RBI with different ROB totals. Also, OOZ plays are not all created equal. Same with in zone. Took many variables.

[edabbs44]

Why don't you start by actually providing specific reasons that support the several strongly declaritive statements you've made in this thread?

We've had multiple conversations on this issue over the years so I'm confident you can find the reasons for my argument among the thousands of words I've authored in the archives on this subject out of respect for those who who either questioned the argument or were curious about the topic.

I'm still struggling to find the reasons behind your strong opinions on this issue. Please point us to them.

http://itsaboutthemoney.net/archives...r-the-new-rbi/

Just found this, a lot of this is in line with my rationale.

[jojo]

http://itsaboutthemoney.net/archives...r-the-new-rbi/

Just found this, a lot of this is in line with my rationale.

Please pick out one or two points you find particularly relevant and lets discuss them in detail.

[edabbs44]

Please pick out one or two points you find particularly relevant and lets discuss them in detail.

This is not to say that UZR is useless, just that is unreliable in single season increments and that unreliability is passed on to WAR, which we habitually use/misuse when discussing single seasons and partial seasons.

[edabbs44]

Please pick out one or two points you find particularly relevant and lets discuss them in detail.

WAR’s move to the mainstream is deeply tied to the rising popularity of FanGraphs. One of the first of it’s “unlikely results” to spark considerable conversation was Ben Zobrist leading AL batters (and finishing behind only Albert Pujols and Zack Greinke overall) in 2009. Zobrist had a breakout season which was impressive by any measure, but his WAR was given a major boost by his defense (only Franklin Gutierrez and Nyjer Morgan got a greater advantage from fielding).

On one level, this seemed legit. Zobrist appeared at every position on the diamond in ’09 and over the years has proven himself to be an above-average defender at second base and in right field. Managers have long lauded the value of versatility and lavished praise on players like Zobrist, Mark DeRosa, and Placido Polanco, who play several key positions well and also swing decent sticks. Zobrist’s looked like evidence of their wisdom.

But while it isn’t much of a stretch to believe that Zobrist’s glove was worth a couple wins to the Rays in 2009, try selling this: According to WAR, in 2011, Carlos Lee has had as much defensive value as Troy Tulowitzki.

[westofyou]

No.

Maybe Henry Chadwick can spread some new light on batting average too

[jojo]

WAR’s move to the mainstream is deeply tied to the rising popularity of FanGraphs. One of the first of it’s “unlikely results” to spark considerable conversation was Ben Zobrist leading AL batters (and finishing behind only Albert Pujols and Zack Greinke overall) in 2009. Zobrist had a breakout season which was impressive by any measure, but his WAR was given a major boost by his defense (only Franklin Gutierrez and Nyjer Morgan got a greater advantage from fielding).

On one level, this seemed legit. Zobrist appeared at every position on the diamond in ’09 and over the years has proven himself to be an above-average defender at second base and in right field. Managers have long lauded the value of versatility and lavished praise on players like Zobrist, Mark DeRosa, and Placido Polanco, who play several key positions well and also swing decent sticks. Zobrist’s looked like evidence of their wisdom.

But while it isn’t much of a stretch to believe that Zobrist’s glove was worth a couple wins to the Rays in 2009, try selling this: According to WAR, in 2011, Carlos Lee has had as much defensive value as Troy Tulowitzki.

Hippeaux was kinda forced to concede that Zobrist's 2009 may have been reasonable given Zobrist has consistently graded out as a plus defender.

When looking for outliers, I get why he picked Lee's 2011 given it was Lee's best defensive year in over half a decade. However, Hippeaux actually demonstrates a fundamentally flawed understanding of WAR when he argues WAR indicates Carlos Lee had as much defensive value as Troy Tulowitzki in 2011. In reality WAR indicates that Tulo was about 1.2 WAR more valuable than Lee solely given the defensive production of each player in 2011. Lee may have been rated as a +11.2 defender across two positions (LF and 1b) but he also had to give most it back because the two positions he played were on the far right of the defensive spectrum (dinged big time with a position adjustment). Meanwhile, Tulo put up his +7.3 at one of the most demanding positions and he also gets credit with his position adjustment. Hippeaux fails to acknowledge this truth about WAR and it's a pretty BIG thing to leave out of his argument. Hippeaux is absolutely incorrect when he states that WAR indicated that Lee had more defensive value than Tulo in 2011. It just simply isn't accurate to argue such a thing.

What's more, Hippeaux inexplicably picked Lee to make his point when in fact Lee is a perfect example of why Hippeaux's point is invalid. Lee's 2011 UZR was based upon split time between first and LF which means estimates of Lee's production relied upon barely 600 defensive innings at each position that season. Hippeaux did this presumably to personify his railing about the uncertainty associated in such UZR sample sizes in the paragraphs above even stating "UZR results get weirder the smaller the sample gets". He clearly knew better which makes his failure to use a caveat when comparing Lee to Tulo that year even more damning to his overall conclusion. WAR doesn't fail to distinguish between players that play many positions well and those who are given it because they play no position well. It dinged Lee pretty harshly for being a lesser of two evils and it pats middle infielders on the back.

Did Lee have a great defensive season in 2011 or was his numbers more random mirage than real? We don't know for sure. He could've had a great year. But it's a year that really sticks out from "his crowd" and there is at least a compelling reason to pause.

If you're arguing that people misapply WAR when they blindly throw it out in conversation, then I'd agree, that can happen. That's the story with any stat however. When using WAR just like any other stat, one has to be aware of it's limitations. Just like Dunn in 2008, Lee's 2011 WAR value is less certain due to sampling issues and the chance that randomness might have skewed things increases. But that's not a damnation of WAR per se....it's a criticism of the misapplication of WAR, i.e. a failure to acknowledge the potential limitation of a specific player's WAR. Hippeaux however does not make a good case for his argument that WAR needs to be thrown out with the bath water.

[MrRedLegger]

If the "Replacement" player could be defined by quantified/calculated stats, somehow, then the argument on WAR would be pretty moot.

A lot of times WAR is used to compare two or more players, but it seems that the standard for comparison is purely mythical...which leads to large debates on the subject.

[dougdirt]

If the "Replacement" player could be defined by quantified/calculated stats, somehow, then the argument on WAR would be pretty moot.

A lot of times WAR is used to compare two or more players, but it seems that the standard for comparison is purely mythical...which leads to large debates on the subject.

Even if such a player is mythical, he is the same mythical player for every comparison in a given year.

[RedsManRick]

The anecdotal link between FB% and high/low UZRs is interesting. I would love to see it analyzed more fully.

But this this article clearly starts with its conclusions, asserting that certain players must be good/poor defenders and being offended when a single season of UZR disagrees. It doesn't allow for the possibility that single season WAR outliers are simply the result of outlier performances (as opposed to bad measurement) nor that teams with high FB% have high UZR OFs because they knew they had a high FB% and managed their defense accordingly.

It's a fancy version of this wonderful argument:
- I have a belief.
- You have presented data that suggests my belief may be wrong.
- I am right and have found some anecdotes that support my position.
- Therefore you are almost certainly wrong.

And this passage makes me wonder if they realize that replacement levels are calculated based on the amount of time the player spent at each position, not based on one of their positions:

There are two types of utilitymen, those who are given the job because they play many positions well and those who are given it because they play no position well. As As yet, WAR struggles to distinguish between the two. It reads Houston’s inability to decide where Lee hurts them least as evidence of Lee’s versatility. It suggests that Howie Kendrick‘s defense at second base has gone from average to exceptional since Mike Scioscia started giving him more starts in left field.

As far as I know, UZR doesn't give bonus points for versatility, it simply grades you based on where you're playing...

And the WAR hates sluggers argument...

However, one can’t help but notice that a cross-section of the most intimidating hitters in the game are treated with relative disdain by the metric. It doesn’t like them because they play first base or left field (or DH), which aren’t scarcity positions. It doesn’t like that they are fat and slow.

Don't forget, it thinks they look funny and have bad breath. That's what it automatically subtracts 2 wins. It doesn't like sluggers who play crappy defense. So what? Why is that wrong? Are those guys supposed to get a pass for playing crap defense (or none at all)? No, we just rely on the magic of the player interaction idea; that guys like Prince Fielder massively boost the performance of the guys around them, despite the fact that it has been studied and found that the effect is quite small.

And then the lovely final argument; people use WAR wrong. Therefore it sucks. Sure, anybody who is getting on their high horse about a win's worth of difference should be knocked down a peg. But I'd still like to hear a better alternative. Because when the alternative is ignoring defense entirely or just giving it lip service as a consideration on par with whether the guy smiles nice of for the cameras, I'll still with the flawed stat.

[edabbs44]

Apologies for the non-specific response as it is very difficult to respond to large posts on an iPhone but I guess I have one question outstanding on the topic, or one string of questions.

When did UZR move from being unreliable in small samples to now being a proper measurement in those same smaller samples? And, to jojo and RMR, do you really believe that players like Bourn, Bruce, Nyjer and others have their true defensive values fluctuate by 2, 3 or more than 3 wins per season?

[RedsManRick]

Apologies for the non-specific response as it is very difficult to respond to large posts on an iPhone but I guess I have one question outstanding on the topic, or one string of questions.

When did UZR move from being unreliable in small samples to now being a proper measurement in those same smaller samples? And, to jojo and RMR, do you really believe that players like Bourn, Bruce, Nyjer and others have their true defensive values fluctuate by 2, 3 or more than 3 wins per season?

"Players like them"? I don't think it's a function of player type. I think a few players had big swings of performance. Is it possible, maybe even likely, that UZR overstated the swings in some cases? Sure. But I'm also willing to accept the possibility of swings ghat large at the extremes. I'm not going to dismiss is out of hand just because it doesn't feel right. Players experience very large swings in offensive performance all the time. It's not out of the realm of possibility that a guy has a great defensive year followed by a mediocre one and that over 140+ games that adds up to 20 runs.

But again, if I'm trying to get my head around players overall value, what's your proposed alternative? Fully accepting that UZR has a fair amount of measurement error, show me the better way.

[mdccclxix]

FWIW, I haven't seen WAR used much around here in a long while. It seemed to be a better go-to stat when all of us were trying to "fix" what was wrong with the Reds and applying all sorts of scenarios to the future. Free agency, trades, projections, etc. It was a simple overall stat to apply to a complex rebuilding process that had no certain future.

The result of that has been a pretty decent familiarity with WAR, but with that familiarity has come rarity in usage due to its, in my opinion, damning shortcomings.

First you have the fWAR, rWAR, VORP (no one uses VORP though) to contend with clarifying. That alone separates it from another criticized stat like batting average because batting average is batting average everywhere. Not many here are afraid of the "nuance", if that's what it can be lightly called, of the different WAR's, but it does get tiresome and just un-stat-like. You can't just lay it out and have it stand on it's own. You have to sell it a little bit, or hope the buyer knows what they're getting.

Secondly, the defensive side of it has been acknowledged over and over for being imperfect. At some point a few years ago I thought we'd all just wait for the advanced computer metrics on defense before we'd even try to sort out UZR or any defensive stat. It's a long and difficult task to read the defensive stats accurately. It's like sailing the ocean in the 1600's. You might land on the right continent, you might not, but it's probably been a learning experience.

For all that, I just look at the offensive side of WAR. In fact, I find myself preferring the espn number on that. Oh well.

[edabbs44]

"Players like them"? I don't think it's a function of player type. I think a few players had big swings of performance. Is it possible, maybe even likely, that UZR overstated the swings in some cases? Sure. But I'm also willing to accept the possibility of swings ghat large at the extremes. I'm not going to dismiss is out of hand just because it doesn't feel right. Players experience very large swings in offensive performance all the time. It's not out of the realm of possibility that a guy has a great defensive year followed by a mediocre one and that over 140+ games that adds up to 20 runs.

But again, if I'm trying to get my head around players overall value, what's your proposed alternative? Fully accepting that UZR has a fair amount of measurement error, show me the better way.

Do you see the same sized swings in offensive players from year to year w/o an obvious type of reason? Look at Bourn, he goes from #4 in the majors for OFers in 2010 to #50 in 2011 to #1 in 2012. Kemp from 2009-2011, 22nd to dead last (62nd) to 37th. Crawford is God's gift until he gets to Fenway and then all of a sudden is a liability. Nyjer Morgan goes from an UZR/150 of 34.7 in 2009 (and a WAR of 5.2, #7 in all of baseball) to an UZR/150 of 2.3 in 2010, knocking his WAR down to 1.0 the following season.

And, even in those instances where someone may have an off year at the plate, it is much easier to see that and prove that compared to an abstract metric like UZR. If there were a way to illustrate that these material swings in defensive performance were legitimate, I would obviously feel better about using UZR and related metrics.

And maybe the answer to your question about getting your head around overall value is that the industry isn't there yet. So you can obviously use what you wish, but excuse some if WAR isn't used as a legitimate measurement tool.