It ultimately comes down to a highly-subjective judgment of Mauer’s defense, and I don’t think we can justifiably claim that his defense is elite since he allowed the second-most wild pitches/passed balls in the AL and had the lowest range factor among qualified catchers.

Admittedly, I don't know much about defensive stats, but how important could range factor possibly be for a catcher?

SIX MORE WINS?

how important could range factor possibly be for a catcher?

Not very. It's hard to judge a catcher's defense. Based on what we have available, I don't think we can say that Mauer's defense is good enough to close the slight WAR edge Zobrist has over him.

SIX MORE WINS?

Chyeah. In '09, Figgins was worth 6.8 WAR; Feliz was worth 0.8. That factors in my calculation of non-base-stealing base running from Baseball Prospectus though, so I could have screwed up somewhere.

Wow. How does it come down to Mauers defense? Leading the league in OBP and SLG wasn't enough? He's not a DH ... sheesh.

How does it come down to Mauers defense?

Zobrist: 8.6 WAR
Mauer: 8.2 WAR

That's from FanGraphs, which doesn't count base running (Zobrist was better) and catcher's defense since it's so hard to quantify. In order for Mauer to be statistically more valuable, he'd have to be a pretty good defender and I don't think there's enough evidence to support that.

I am not on board with giving Mauer zero credit for his glove, while giving Zobrist a huge bonus for his fielding.

It's not giving him zero credit; it's simply stating that Mauer's defense likely isn't good enough to make up for the difference in value. That's why I said it comes down to his defense. If he is an awesome defensive catcher, he's the MVP. If not, he's not. :)

It's not giving him zero credit; it's simply stating that Mauer's defense likely isn't good enough to make up for the difference in value. That's why I said it comes down to his defense. If he is an awesome defensive catcher, he's the MVP. If not, he's not. :)

Serious question: What credit are you giving him for his defense?

It's not giving him zero credit; it's simply stating that Mauer's defense likely isn't good enough to make up for the difference in value. That's why I said it comes down to his defense. If he is an awesome defensive catcher, he's the MVP. If not, he's not. :)

But Zobrist's defense has substantial SSS issues, being split across several positions. He may have had disproportionately easy to field chances.

I think you want to discount that value pretty heavily compared to offensive value.

IMHO Mauer is the clear MVP.

I am not overly familiar with Fangraphs or WAR but the fielding column for Mauer is blank. Does that figure into the calculations at all?

I am not overly familiar with Fangraphs or WAR but the fielding column for Mauer is blank. Does that figure into the calculations at all?

They say they don't have the ability to rate catcher defense, so everyone is assumed to be average.

Thanks Snapper. I guess I'm back to SOSH's question: what credit are you giving Mauer for his defense?

Also, Fangraphs has Zobrist as the 3rd best defender in MLB, behind Guitterez and Morgan ... an awful big reach due to SSS at any one position.

Like Brian said (or paraphrased) Zobrist is a GG at nine positions, Mauer is a slug. Case closed.

But Zobrist's defense has substantial SSS issues, being split across several positions.

If we're trying to evaluate how good Zobrist was with the glove in 2009, why does his time being split between different positions hurt him?

And expecting Figgins to repeat his 2009 WAR is insane.

If we're trying to evaluate how good Zobrist was with the glove in 2009, why does his time being split between different positions hurt him?

Because, in a small sample it is less likely that his chances will be of average difficulty.

UZR rates a fielder vs the average likelihood of a play being made on a hit-type (ground, fly, liner, fliner) in a zone. But some balls in a given type/zone combo are harder to field than others (velocity, location in the zone, etc.).
Over a large sample size, we assume this averages out.

But, one-year is not a large sample for defense (about the same # of observations as 2 months of hitting. Less than 100 Gs (as is true for each of Zobrist's positions) is a tiny sample.

So, there is a huge error band around Zobrist's defensive contributions that doesn't exist for Mauer's hitting.

Mauer is a solid defensive catcher if not beyond that and to think anyone else in the American League is the MVP is to abdicate one's claim on having common sense.

Ben Zobrist? Please. This entire issue undermines the credibility of our community when we put forth the notion that a utility man having a career year is the MVP. We should take Zobrist's numbers as a sign that even the best of assessment methods need to be monitored for peculiar outputs not as gospel.

Folks have doubts about Mauer's defense but take everything on Zobrist as known?

So, there is a huge error band around Zobrist's defensive contributions that doesn't exist for Mauer's hitting.

And maybe we can be more forgiving of the margin for error if anecdotal evidence suggests that Zobrist looked like Roberto Clemente in right and like Chico Lind at second. If there's such evidence, I'm unaware of it.

Crashburn, I expect better. Citing UZR small samples and catcher range factors chapter and verse, then valuing Chone Figgins based on a 31-year old season that is about twice as good as any of his previous years?

Mauer is a solid defensive catcher if not beyond that and to think anyone else in the American League is the MVP is to abdicate one's claim on having common sense.

Even if he was below average defensively your point about the MVP stands.

And maybe we can be more forgiving of the margin for error if anecdotal evidence suggests that Zobrist looked like Roberto Clemente in right and like Chico Lind at second.

Yes. It seems strange, to say the least, that Zobrist would put up an all-time great performance (at two separate positions) and no one would particularly notcie.

It ultimately comes down to a highly-subjective judgment of Mauer’s defense, and I don’t think we can justifiably claim that his defense is elite since he allowed the second-most wild pitches/passed balls in the AL and had the lowest range factor among qualified catchers.

First, far, far too much confidence in highly flawed and highly variable defensive stats. Second, Zobrist gets 2.6 WAR from defense at multiple positions including 68 games at a corner OF position, yet Mauer would have to be an "elite" defensive catcher to make up the .4 difference in overall WAR? Umm no.

That's from FanGraphs, which doesn't count base running (Zobrist was better) and catcher's defense since it's so hard to quantify. In order for Mauer to be statistically more valuable, he'd have to be a pretty good defender and I don't think there's enough evidence to support that.

I'd like to pile on and call ######## on this. In order to make Zobrist more valuable you have to have to selectively choose to believe all the numbers that favour him whiel selectively ignoring all the factors that don't. Nice.

That factors in my calculation of non-base-stealing base running from Baseball Prospectus though, so I could have screwed up somewhere.

It also assumes that the 19 UZR/150 Figgins put up at 3B this year is his true talent level. I am doubtful of that.

Because, in a small sample it is less likely that his chances will be of average difficulty.

UZR rates a fielder vs the average likelihood of a play being made on a hit-type (ground, fly, liner, fliner) in a zone. But some balls in a given type/zone combo are harder to field than others (velocity, location in the zone, etc.).
Over a large sample size, we assume this averages out.

But, one-year is not a large sample for defense (about the same # of observations as 2 months of hitting. Less than 100 Gs (as is true for each of Zobrist's positions) is a tiny sample.

So, there is a huge error band around Zobrist's defensive contributions that doesn't exist for Mauer's hitting.

I'm not questioning the value of offensive metrics versus defensive metrics. Rather, I'm questioning why Zobrist spending time at several different positions rather than staying at the same position makes him less of an MVP candidate. Yes, the defensive stats are a very small sample at each position. But all of those small samples add up to a less-small sample, one which is just as large as any other player's single season fielding sample.

I'm not questioning the value of offensive metrics versus defensive metrics. Rather, I'm questioning why Zobrist spending time at several different positions rather than staying at the same position makes him less of an MVP candidate.

I think the argument is that the numbers, divided as they are between three positions, seem even less reliable than a full-season's worth at one position.* If snapper had complete faith in the defensive numbers, I don't think he'd have a problem with Zobrist's compiling them at multiple positions.

* This seems true to me as well, but I have no idea if the math supports it. Is one small sample of data more reliable than three small samples, if they both have the same number of (in this case) chances.

Yeah, I don't see how you can justify that Figgins 2009 = Figgins going forward in any reasonable statistical way.

He had 729 PAs, compared to an average of around 570 in 2006-2009. Is there some reason to think a 31 year old Figgins has established a new health skill that will carry forward?

He's got a career OBP of .363. Again, why should we think that his .395 2009 OBP is his new talent level and completely ignore the prior 3346 PAs of his career? Sure, weigh 2009 more heavily, but that still doesn't get him to a projected .395 OBP in 2010.

And as far as his defense, I have a hard time thinking someone who's got a career UZR/150 of +8 at 3B is now all of a sudden a +17 defender because that's what he did in 2009, especially at age 32 when defense should be declining.

Baserunning, sure, I can see him maintaining similar value on a rate basis, although once you re-project him with what his OBP should be he won't be as valuable overall because he probably won't be getting on base as much.

If I had to project him right now, I'd probably peg him around .290/.370/.380, with a wOBA of about .340 (compared to this year's .354). At 600 PAs, that's a difference of around 7 runs at the plate from 2009 (9 if you use 729 PAs instead). I'd also project his defense around +10. So even if he's just as valuable in baserunning, both SB and non-SB, you should figure he's going to be around 16 runs less valuable on offense and defense.

That's still a very good player, but that's not a 6.8 WAR player. More like a 5.2 WAR player.

Brian: I guess I'm back to SOSH's question: what credit are you giving Mauer for his defense?

I'm not creating any sort of numerical value for it because people much smarter than me haven't been able to do it effectively thus far. Like I said, it's highly subjective, but I try to base it on what statistics we do have available. Those statistics say that Mauer isn't good at blocking balls in the dirt and he throws out an average amount of base runners (25%). Those are the most important aspects of catching, defensively speaking, and Mauer doesn't rate among the elite in those areas. He could very well call a hell of a game for his pitchers, among other items that we can't quantify accurately yet. But based on the information available, I don't know how you can confidently state that Joe Mauer is an elite defensive catcher.

EDIT: And yes, citing range factor was a bad idea on my part. I didn't want to bog up the interview by repeating verbatim what I had written in my BBA/BDD post.

Harvey: This entire issue undermines the credibility of our community when we put forth the notion that a utility man having a career year is the MVP.

It doesn't matter if he is a "utility man having a career year". The MVP is the Most Valuable Player, and Zobrist was the Most Valuable Player in the American League. If Eric Bruntlett put up a .948 OPS in 2010, he would also be in contention for the MVP. It's not about if his contributions were out of character and unable to be repeated; it's what he did and how it ranks among his peers for the year.

Folks have doubts about Mauer's defense but take everything on Zobrist as known?

It's much easier to gauge defense elsewhere; quantifying a catcher's defense has been hard, unless I completely missed the boat somewhere.

Danny: And expecting Figgins to repeat his 2009 WAR is insane.

Valid point. But at any rate, even if you heavily regress Figgins (as #25 does), he still rates much better than Beltre, Crede, or any of the other available third basemen on the market.

Bad Doctor: Citing UZR small samples

It's not a terribly small sample. Yes it is small, but I think we're overstating the issue. 60% of his defensive innings came at second base, so it's not like we're dealing with 200 defensive innings at 5 positions. He had 715 innings at second base at about a 31 UZR/150. Even if he had a friendly selection of ground balls in that small sample, he's still well above average any way you look at it.

By the way, I say all this in agreement with the sentiment that UZR is not wholly reliable. There's an interesting discussion at Beyond the Box Score on it (link).

Gaelan: In order to make Zobrist more valuable you have to have to selectively choose to believe all the numbers that favour him whiel selectively ignoring all the factors that don't.

This is disingenuous: I acknowledge that Mauer is a better hitter, but Zobrist had a damn fine offensive year and got some bonus points for his speed and defense.

--

Sorry for the delay in responding, real life calls. Good discussion, thanks. :)

The MVP is the Most Valuable Player, and Zobrist was the Most Valuable Player in the American League.

Only if you put 100% faith in the defensive numbers on Fangraphs(mistake #1) and give Mauer a 0 for defense (mistake #2). Mauer doesn't have to be an "elite" defensive catcher to be more valuable than Zobrist. He has to be marginally above average and that's only if you actually accept the values for Zobrist's defense on Fangraphs.

I think the argument is that the numbers, divided as they are between three positions, seem even less reliable than a full-season's worth at one position.* If snapper had complete faith in the defensive numbers, I don't think he'd have a problem with Zobrist's compiling them at multiple positions.

* This seems true to me as well, but I have no idea if the math supports it. Is one small sample of data more reliable than three small samples, if they both have the same number of (in this case) chances.

Correct.

The issue is the number of chances of each type (grounder, fly, liner, fliner) in each zone.

For UZR to work, you have to assume all the chance of a type in a zone are equal. Of course they are not, some fliners are close to liners, some are close to flys, some balls are harder hit, and they are hit to different parts of the zone.

With enough chances, this will average out. But, with Zobrist playing multiple positions, his plays are spread across more zones of the field, lowering the chances that things average out.

I'm guessing Zobrist had a lot of easily fielded ball relative to the average player at his positions. Otherwise, it is hard to reconcile him playing truly otherwordly defense (+40 R/150 at 2B and RF) without people noticing and saying "Wow this guy is Roberto Clemente and Joe Morgan all rolled into one".

Only if you put 100% faith in the defensive numbers on Fangraphs(mistake #1) and give Mauer a 0 for defense (mistake #2).

Not true. Like I said, I acknowledge that UZR isn't infallible; far from it (see the link to BTB). And like I said -- and people keep putting words in my mouth -- I am not failing to give credit to Mauer for defense; I'm just not assigning a numerical value to it as has been done for Zobrist because catcher defense is incredibly hard to quantify. It hasn't been effectively done to date and I sure won't be the first to forge through that frontier, if you catch my drift.

Mauer doesn't have to be an "elite" defensive catcher to be more valuable than Zobrist. He has to be marginally above average

I don't think there's enough evidence to say that he's even above-average. He's average in throwing runners out and he doesn't block balls in the dirt well. Where else can you go to make a case for Mauer defensively?

and that's only if you actually accept the values for Zobrist's defense on Fangraphs.

As I said above, "Even if he had a friendly selection of ground balls in that small sample, he's still well above average any way you look at it."

But, with Zobrist playing multiple positions, his plays are spread across more zones of the field, lowering the chances that things average out.

Why does that make it less likely to even out?

You seem to be employing the gambler's fallacy. You're assuming that if Zobrist stuck at one position, the easily fieldable balls he got at the start of the season would likely be balanced out with more difficult plays later in the year. If he moves around positions, he can somehow avoid this evening out process.

But there's no reason to expect that type of balancing out; those easily fielded balls are already in the books. In fact, a skewed distribution of chances may be due to a tendency of the pitching staff, which would suggest sticking at one position would make Zobrist more likely to be the season-long beneficiary of easy plays.

Saying each position sample is too small seems tantamount to dicing his season up into months and then tossing away all the data because each month's sample is too small to draw any conclusions from.

As I said above, "Even if he had a friendly selection of ground balls in that small sample, he's still well above average any way you look at it."

OK, so deduct 5 runs from his +26 on defense. Then he is still "well above average" as you put it...but wait, then he is at 8.1 WAR and behind Mauer. Simply put, I am far, far more confident that Mauer was actually worth that 8.2 WAR than I am that Zobrist was worth that 8.6. The error margins on Zobrist's value are far too great.

As I said above, "Even if he had a friendly selection of ground balls in that small sample, he's still well above average any way you look at it."

Yes, but he's getting 2.6 WAR of his value from defense, and he only leads Mauer by 0.4 WAR.

His defensive rating at 2B and RF are incredible, +40 runs/150 Gs, when the absolute best fielders are no more than +20 R/150 on a consistent basis.

Put it this way, if Zobrist really was an elite 2B/RF, but normal elite, not +40, he loses about 1.3 WAR of value, and falls behind Mauer 8.2 to 7.3, even giving Mauer nothing for D.

Zobrist's UZR #'s are just not credible. They assume he played like one of the best fielders ever at two separate positions.

Edit: half a coke to ryan :-)

Edit: half a coke to ryan :-)

Ouch. Mistaken for a Canadian. Now that is insulting.

It's much easier to gauge defense elsewhere;

The point is it's a small sample of this and a small sample of that. It's not a question of easy and hard plays evening out over time. It's a question of errors of description in the sample itself.

It's a question of errors of description in the sample itself.

Right, I know it's a small sample but I think it's being overstated. 715 innings is not 1200 innings, of course, but it's also not 200 innings. Can we really say that if he had his other 500 innings at second base that his UZR would go from 31 to, I don't know, 15?

Chase Utley was +46 in Dewan's Fielding Bible last year. No way he's that good but he did that in about 1400 innings. We're just as likely to throw that out as a fluke despite the sample size being twice as large as Zobrist's. Sometimes fielders just have really good years. Whether it's fluky or not is not something that should be counted against a player; he did what he did.

If Zobrist reall were this fantastic on defense he wouldn't have had to wait this long to get on the field nor would he be floated around the diamond.
There numerous reasons to look askance at Zobrist's defensive output as being curious. To declare "Yes, he is great" based on such limited info is more curious still.

If this communite wants real credibility better to fight for the guy with both the seasonal performance and long term track record STILL struggling to get his due. Batting one's eyelashes at the out of nowhere guy with the uber cool nerd stats and nothing else.

And I am a long-time nerd so nobody get their britches in a bunch

If Zobrist reall were this fantastic on defense he wouldn't have had to wait this long to get on the field nor would he be floated around the diamond.

You're not arguing against Zobrist's value; you're arguing against Zobrist's future as a player. I agree that Zobrist will not be as good as he was in 2009, but it doesn't discount how ridiculously good he was in 2009.

Track record has nothing to do with MVP.

Sigh. I am pointing out Zobrist's alleged defensive prowess should be deemed curious.

The metrics are not infallible. One should look at ALL the evidence.

Where is the thread touting Zobrist's amazing defense? What blogger has screamed for others to watch this guy play in the field?

Sigh. I am pointing out Zobrist's alleged defensive prowess should be deemed curious.

The metrics are not infallible. One should look at ALL the evidence.

100% Harveys.

I do quantitative analysis for a living. When the data or the model contradict a analyst's sound judgment, the model is wrong 98% of the time.

At least once every few months one of my team will publish a risk report showing VaR (a risk statistic) going from $5M to $20M in one day. Every time I see it, I immediately say it's wrong, and ask how could they publish such a wacky number.

Every time they come back with some story about market volatility and changing positions, and every time when I make them check, there's an error.

You can't let a model supercede sound judgment and common sense. That's how we got AAA bonds based on CCC loans, which could never, ever default, but are now trading at $0.10 on the dollar.

It is possible that Ben Zobrist played like the greatest RF and greatest 2B-man ever this season, but is it likely? Is it likely that he did this and no one commented on such a great performance, or noticed his defensive ability before this season?

Ouch. Mistaken for a Canadian. Now that is insulting.

It's not easy keeping up with the Joneses.

Ouch. Mistaken for a Canadian. Now that is insulting.

My sincere apologies. One of you needs to add a wacky parenthetical to your handle.

It's not giving him zero credit; it's simply stating that Mauer's defense likely isn't good enough to make up for the difference in value. That's why I said it comes down to his defense. If he is an awesome defensive catcher, he's the MVP. If not, he's not. :)

But it doesn't come down to defense. It comes down to WAR, with defense and anything else as a tie-breaker. Right?

You're phrasing this as if we should all accept WAR, and the only factors worth talking about are those that WAR may be missing. The premise is laughable.

The MVP is the Most Valuable Player, and Zobrist was the Most Valuable Player in the American League.

The problem is that the numbers don't show this. Even if we take UZR as gospel Zobrist's numbers have to be regressed not only with reference to future projections but also with regard actual 2009 performance. This is a fact that is beyond dispute. Ergo your usage of the numbers is not supported by the numbers themselves.

I'm reading MGL's explanation, and with his more in-depth explanation, it turns out I am wrong to count Zobrist's defense so heavily. I rescind my argument... for now.

Sorry this turned into 15-on-1 but as Snapper said, when you get a wild outlier of an answer it's worth checking.