The Increasing Over Reliance on WAR

[nsacpi]

Right. And if you tried, you'd have a whole 'nother layer of subjectivity.

It doesn't have to be subjective. We have data from Simmons rookie season on what happened to balls hit to the left side of the infield. Believe me those data show an enormous improvement in our infield defense after Simmons became a starter. An improvement that went well beyond what was being captured by Simmons' defensive stats.

 

[thewupk]

Sure. But the park with a deep RCF will make it more possible to make exceptional plays. So it cuts both ways, doesn't it?

And I've heard the "guy next to this guy is good, so this guy looks better" argument a lot. If a great CF ranges into RCF and takes a ball that the RF would have possibly gotten, how does that make the RFs defensive metrics look better? I don't think it does. The team is better off, but not the RFs defensive rating.

And in the infield, similar observation - "CJ looks better on the metrics because Andrelton plays next to him..." No, he doesn't. CJ gets to the ball or he doesnt. There will be a few to the left that he doesn't get to that Andrelton will and that will improve the team's defense. But not CJs.

I've done some research on the matter at least in regards to the Braves. While Andruw was here his corner outfield mates posted the best UZR numbers or their careers. Could be just be noise since it's based on one player but I feel that would be the case with any great defender in center. Yes the RCF and LCF chances would be taken away. But they would presumably get to balls closer to the line, etc.

In regards to CJ his UZR/150 has been in the double digit negatives while playing on the Dbacks and Astros. It's been at -8 and -4 the past two years. So there is some improvement there. How much of that is due to Andrelton? Who knows.

 

[GovClintonTyree]

It doesn't have to be subjective. We have data from Simmons rookie season on what happened to balls hit to the left side of the infield. Believe me those data show an enormous improvement in our infield defense after Simmons became a starter. An improvement that went well beyond what was being captured by Simmons' defensive stats.

...and so you're going to attribute that to Simmons' impact. What if CJ actually had (shocking, but possible) more range than Chipper?

See what I mean? Somebody's going to make a judgement.

 

[GovClintonTyree]

I don't think WAR is predicated upon the concept that defense is as important as offense. It is predicated on the idea that run prevention is as important as run production. But a major part of run prevention is pitching. So the defensive component's overall value is the value of run prevention minus the contribution that pitching makes to run prevention. And this is in part why a stat like FIP (fielding independent pitching) fits conceptually.

I understand that it is predicated on run creation and run prevention. And certainly pitching is a (the biggest?) component of run prevention, and I like the way you're using FIP to isolate that variable.

NOW what I am suggesting is that the delta between the best fielder at a position and the worst is less than the delta between hitters - as well as being very difficult to quantify.

Does that help?

 

[thewupk]

I don't think WAR is predicated upon the concept that defense is as important as offense. It is predicated on the idea that run prevention is as important as run production. But a major part of run prevention is pitching. So the defensive component's overall value is the value of run prevention minus the contribution that pitching makes to run prevention. And this is in part why a stat like FIP (fielding independent pitching) fits conceptually.

That and an elite hitter will have a 50+ RAR (runs above replacement) while an elite defensive player will have around 20 RAR. Some of that is due to the chances you have on defense compared to offense and why you really need 3 years of data for defense. So why you can say that WAR makes defense and offense the same it's not really. An elite hitter with average defense is likely to be 50+ RAR. An average hitter with elite defense is likely going to be 20+ rar. That's without looking into base running, positional adjustment, playing time, etc.

 

[chop2chip]

I usually refer to WAR myself most of the time when I want to get a quick surface evaluation of a player and how "valuable" (very arbitrary). But I think there are real limitations that I think need to be acknowledged. For example, two years ago Dan Uggla was rated as a plus defender and subsequently had a WAR of 3.3. However, there are mountains of evidence that suggest he is otherwise a very poor defender which inflated his total quite substantially. I think public defensive data is fine in the aggregate but individual samples of player data can be misleading.

My point in the end is that I think it's fine to use it as a pretty effective rough estimate, but I also feel like it is imprecise enough to question in individual cases when there is a logical reason. For a player like Markakis, thats extremely important since his WAR is pretty dependent on his perceived range. We just don't have the slightest idea about how defensive performance is year over year. Even markakis, he was an exceptional defender based off the metrics for the first 4 seasons of his career and then the next year he cratered and he's been bad ever since, without major injury. It's just a weird phenomenon.

 

[thewupk]

...and so you're going to attribute that to Simmons' impact. What if CJ actually had (shocking, but possible) more range than Chipper?

See what I mean? Somebody's going to make a judgement.

I would hope CJ in his prime would have more range than a 40 year old broken down Chipper.

 

[thewupk]

I usually refer to WAR myself most of the time when I want to get a quick surface evaluation of a player and how "valuable" (very arbitrary). But I think there are real limitations that I think need to be acknowledged. For example, two years ago Dan Uggla was rated as a plus defender and subsequently had a WAR of 3.3. However, there are mountains of evidence that suggest he is otherwise a very poor defender which inflated his total quite substantially. I think public defensive data is fine in the aggregate but individual samples of player data can be misleading.

My point in the end is that I think it's fine to use it as a pretty effective rough estimate, but I also feel like it is imprecise enough to question in individual cases when there is a logical reason. For a player like Markakis, thats extremely important since his WAR is pretty dependent on his perceived range. We just don't have the slightest idea about how defensive performance is year over year. Even markakis, he was an exceptional defender based off the metrics for the first 4 seasons of his career and then the next year he cratered and he's been bad ever since, without major injury. It's just a weird phenomenon.

I don't think it's that odd for a player to have a better year in a given season (the Uggla example). To me that is what happened and should be presented as such. However going forward I wouldn't expect that to still be the case and as we saw the next season he performed at his expected level. Just like bad hitters can get hot for a month or two at a time. Poor defenders can be 'good' for a year.

 

[GovClintonTyree]

I don't think WAR is predicated upon the concept that defense is as important as offense. It is predicated on the idea that run prevention is as important as run production. But a major part of run prevention is pitching. So the defensive component's overall value is the value of run prevention minus the contribution that pitching makes to run prevention. And this is in part why a stat like FIP (fielding independent pitching) fits conceptually.

The first year I played fantasy football I looked at the points scored under our system and noticed that Team Defense scored a lot of points, almost as many as QBs in our system, and I decided I would be more aggressive in drafting a defense. I got the #1 defense in the second round. My more experienced league mates snickered and cracked on me. Why? I had miscalculated the importance of Team Defense. The best scored 120 and the worst scored 90 and there wasn't a whole lot of difference between one and the other.

Most baseball defense is made up of routine plays. There are only a few chances to make a difference per game or per week. Thus, the delta between the best and the worst defender is small. You don't throw out the routine and judge A vs B. You keep it in there, lest you overvalue the delta between them.

Kevin Towers decided he was going to perform his own little version of Moneyball a few years ago by trading away big hitters with few ancillary skills or intangibles in favor of gritty ball players who were fundamentally sound and played the game "the right way". His results were disastrous, ultimately costing him his job.

The delta in runs created between Justin Upton and Ryan Langerhans is far greater than the delta in runs prevented.

 

[chop2chip]

I don't think it's that odd for a player to have a better year in a given season (the Uggla example). To me that is what happened and should be presented as such. However going forward I wouldn't expect that to still be the case and as we saw the next season he performed at his expected level. Just like bad hitters can get hot for a month or two at a time. Poor defenders can be 'good' for a year.

I agree with the he premise of variation in defense. I just think theoretically there should be much less variation then what the metrics often show. But again that's only theory because at this time I don't believe we have the defensive metrics that reliable to grade. If and when FIELDfx becomes available to the public then I think we are going to uncover a lot of data that will differ with the current measures.

 

[GovClintonTyree]

I would hope CJ in his prime would have more range than a 40 year old broken down Chipper.

Right. My point is, somebody's got to decide how much of the Braves improved left side defense is attributable to which factor. And that's a tough call, and reasonable minds can disagree.

 

[GovClintonTyree]

I agree with the he premise of variation in defense. I just think theoretically there should be much less variation then what the metrics often show. But again that's only theory because at this time I don't believe we have the defensive metrics that reliable to grade. If and when FIELDfx becomes available to the public then I think we are going to uncover a lot of data that will differ with the current measures.

FieldFX. That's it. Couldn't remember that.

I think it will paint a significantly different picture than what we're getting right now and I look forward to it.

 

[zitothebrave]

I agree with the he premise of variation in defense. I just think theoretically there should be much less variation then what the metrics often show. But again that's only theory because at this time I don't believe we have the defensive metrics that reliable to grade. If and when FIELDfx becomes available to the public then I think we are going to uncover a lot of data that will differ with the current measures.

I don't think in many ways field f/x will differ but I think it will hone it more accurately. Instead of a hard hit liner drive, you get the angle and velocity. My hunch is field f/ x will confirm UZR numbers, but refine the output.

 

[zitothebrave]

I don't think it's that odd for a player to have a better year in a given season (the Uggla example). To me that is what happened and should be presented as such. However going forward I wouldn't expect that to still be the case and as we saw the next season he performed at his expected level. Just like bad hitters can get hot for a month or two at a time. Poor defenders can be 'good' for a year.

Correct, the year Uggla rated positively defensively he played much better in the field. He wasn't Ugglaing balls at his normal clip,

 

[GovClintonTyree]

I usually refer to WAR myself most of the time when I want to get a quick surface evaluation of a player and how "valuable" (very arbitrary). But I think there are real limitations that I think need to be acknowledged. For example, two years ago Dan Uggla was rated as a plus defender and subsequently had a WAR of 3.3. However, there are mountains of evidence that suggest he is otherwise a very poor defender which inflated his total quite substantially. I think public defensive data is fine in the aggregate but individual samples of player data can be misleading.

My point in the end is that I think it's fine to use it as a pretty effective rough estimate, but I also feel like it is imprecise enough to question in individual cases when there is a logical reason. For a player like Markakis, thats extremely important since his WAR is pretty dependent on his perceived range. We just don't have the slightest idea about how defensive performance is year over year. Even markakis, he was an exceptional defender based off the metrics for the first 4 seasons of his career and then the next year he cratered and he's been bad ever since, without major injury. It's just a weird phenomenon.

So why not knock down the impact of the volatile component relative to the more objective component?

 

[NinersSBChamps]

Gonna be interesting to see if BRule can figure out how to be logged in with both accounts at the same time.

 

[chop2chip]

I don't think in many ways field f/x will differ but I think it will hone it more accurately. Instead of a hard hit liner drive, you get the angle and velocity. My hunch is field f/ x will confirm UZR numbers, but refine the output.

I agree, but again, in the aggregate. I don't believe UZR to be a total shot in the dark. Only it is something that's likely only accurate over a large sample size.

 

[thewupk]

Right. My point is, somebody's got to decide how much of the Braves improved left side defense is attributable to which factor. And that's a tough call, and reasonable minds can disagree.

I think it's a combination of both. As I said earlier. Chris' defense, while bad, isn't as bad as it looked while on other teams. Can this be attributed to him pretty much playing full time? Having Andrelton to his left? A combination of both?

 

[thewupk]

So why not knock down the impact of the volatile component relative to the more objective component?

As I mentioned earlier. It's not equal in the amount that it applies to WAR. An elite hitter will produce 50+ runs WAR while an elite defender only provides around 20. So based on the term of 'elite'. Hitters are more valuable and I don't think anyone has ever argued against that. I think the problem occurs with some is when you see someone with great offense and ok defense rated the same or below as someone with good offense and elite defense.

 

[chop2chip]

So why not knock down the impact of the volatile component relative to the more objective component?

I think it's fair to keep the same formula. You don't want to interfere with the theory because that would be bad science. I just see it as more of an indictment on the concept of WAR itself.