Baseball for the Thinking Fan

Login | Register | Feedback

btf_logo
You are here > Home > Baseball Newsstand > Baseball Primer Newsblog > Discussion
Baseball Primer Newsblog
— The Best News Links from the Baseball Newsstand

Sunday, November 25, 2012

Don Malcolm: A WATERY BLOODBATH

Malcolm with the final say on the Cabrera/Trout dust-up…or: Baby the Reign Must Fall (doff’s Murray way).

And in the case of the 2012 AL MVP, virtually all of the intrigue in it stems from the fact that one system ranked Mike Trout’s non-offensive value at such an extreme level that it made the comparison with Cabrera into something egregiously lopsided. Our suspicion is that the 28 voters entrusted to decide the AL MVP (and—hey, really? just twenty-eight?? is that not one of the most ridiculously small sample sizes ever for such an important honor???) may well have looked at the various “advanced numbers” as part of their decision-making process. Trout was touted heavily in the MSM while his performance was at its peak from late July-mid August, and at least a smattering of those voters are known to reference those numbers.

...Let’s close with a thought about what would have happened if the hard-sell harping on that outrageously high gap between Trout and Cabrera hadn’t turned away some voters who may have been receptive to shifting their vote. If eight voters had shifted from Cabrera to Trout, we’d wind up in a virtual tie: that vote tally would shift from 362-281 to 322-321. Then, if the inestimably unmentionable Sheldon Ocker of the Akron Beacon Journal hadn’t been the lone wingnut deviation from the Cabrera-Trout binary (either voted 1-2 or 2-1 by 96% of the voting population), we mighta coulda (shoulda?) had a tie in this MVP race, something that traditionalists and the various flavors of Stalinists in the little world of numberology might have found a way to live with—a kind of benevolent whitewash to the watery bloodbath that will now continue to divide rather than unite.

(In any event, we sure as hell would like to have seen a much larger BBWAA voting population involved in this result…with technology and analysis being what it is today, the notion of limiting the voting population to league beat writers and a few other random chimps is as outmoded as the Model T. We wonder what a full BBWAA membership vote on the 2012 MVP might have looked like.)

That day is further off, and while in the real world it’s the “bad guys” who are keeping that from happening, here in this little sub-region of “reality” it’s the “good guys” who are doing same. And the “good guys” should know that it’s much more possible—and much more feasible—to have “co-MVPs” than “co-Presidents.” Keeping that dichotomy in mind might keep things in better perspective.

Repoz Posted: November 25, 2012 at 10:10 AM | 56 comment(s) Login to Bookmark
  Tags: history, sabermetrics

Reader Comments and Retorts

Go to end of page

Statements posted here are those of our readers and do not represent the BaseballThinkFactory. Names are provided by the poster and are not verified. We ask that posters follow our submission policy. Please report any inappropriate comments.

   1. valuearbitrageur Posted: November 25, 2012 at 12:30 PM (#4309123)
Ugh. Yes, you can jump through numerous hoops to try to prove Cabrera the best offensive player, but never by much, and no matter how skepticl you are of Trouts defensive value, you cant deny it was far higher than Cabreras. Trout was easily the MVP even if you believe WAR overstated his value in every way possible.

And even if you are sweetly bearish empty headed on Trout, what about Robby Cano? He was clearly better than Cabrera as well.

Cabrera was the best at slo pitch softball stats this year, terrible at everything else.
   2. BDC Posted: November 25, 2012 at 12:46 PM (#4309131)
Cabrera was the best at slo pitch softball stats this year

Well, he's darn lucky that runs score the same way in major-league baseball as they do in slowpitch :)
   3. Robert in Manhattan Beach Posted: November 25, 2012 at 12:53 PM (#4309136)

Cabrera was the best at slo pitch softball stats this year, terrible at everything else.


And another person who doesn't know what the word 'terrible' means. If he had been truly terrible at these things - ran the bases like Bengie Molina, or played third like Brooks Conrad - he would have lost, but he was passable at both so he didn't lose points in any voters own personal WAR calculation. The only thing he was truly terrible at was hitting into double plays but that isn't on most folks radar.
   4. Avoid running at all times.-S. Paige Posted: November 25, 2012 at 01:03 PM (#4309143)
Was Don Malcolm ever a music critic for one of the NYC alternative weekly rags (Village Voice, NY Press) in the early to mid '90s? His writing has that quality about it.
   5. Matt Welch Posted: November 25, 2012 at 02:27 PM (#4309180)
"Stalinists."
   6. valuearbitrageur Posted: November 25, 2012 at 02:35 PM (#4309184)
And another person who doesn't know what the word 'terrible' means. If he had been truly terrible at these things - ran the bases like Bengie Molina, or played third like Brooks Conrad - he would have lost, but he was passable at both so he didn't lose points in any voters own personal WAR calculation. The only thing he was truly terrible at was hitting into double plays but that isn't on most folks radar.


Another person who doesn't know what the word "passable" means.

Worst third baseman in all of the MLB by UZR, close to bottom of league in base running, and led the league in GIDP. That's not passable at those things, it is legitimately terrible.
   7. Juilin Sandar to Conkling Speedwell (Arjun) Posted: November 25, 2012 at 02:45 PM (#4309188)
I might be the only one, but I genuinely find it very difficult to read Don Malcolm's stuff. I just can't make it past the stylistic parts; it's too much for me.
   8. Danny Posted: November 25, 2012 at 03:21 PM (#4309194)
#slatepitches
   9. puck Posted: November 25, 2012 at 04:11 PM (#4309209)
Worst third baseman in all of the MLB by UZR


An aside: if you set the innings to qualify lower, the worst 3B by UZR is the Rockies' Chris Nelson, a converted middle infielder who split time at 3rd with Jordan Pacheco, a converted 2B/catcher. Those two combined for -21.8, and generally looked about that bad.
   10. Greg K Posted: November 25, 2012 at 07:32 PM (#4309263)
I think it's a testament to Cabrera's hitting that he was one of the best players in the world in 2012 despite bringing not much else to the table.

In a similar vein I've always been fascinated by Mark Reynolds' 2009 season in that it answered the age old question - can you strike out nearly every time you're up and make an error on seemingly ever ball hit to you and still be a useful player?
   11. Cooper Nielson Posted: November 25, 2012 at 10:45 PM (#4309381)
I might be the only one, but I genuinely find it very difficult to read Don Malcolm's stuff. I just can't make it past the stylistic parts; it's too much for me.

I agree, though I found this one more readable than most.

Worst third baseman in all of the MLB by UZR, close to bottom of league in base running, and led the league in GIDP. That's not passable at those things, it is legitimately terrible.

Come on, "worst third baseman in all of the MLB by UZR" doesn't automatically equate to "terrible." Yes, he finished 13th out of 13 in UZR among "qualified 3B," though he was indistinguishable from 12th, and not far from 11th. Granted, no one wants to be last on any list, but this also shows that he was good enough to play every day at 3B. (Drop the inning limit to 100, rank by UZR/150, and Cabrera finishes 63rd of 79; make it 200 innings and he's 50th of 57 -- and comfortably above the truly terrible guys.) If Cabrera's a "legitimately terrible" third baseman, what does that make Michael Young, Mark Reynolds, Edwin Encarnacion, et al.? Is there an even stronger adjective for "bad" that you would use?

Also, how much do you trust UZR? According to UZR, Jhonny Peralta was the 4th best SS in all of MLB this year. The worst two (non-Jeter category) were Asdrubal Cabrera and Alcides Escobar. Maybe that's right, but it seems awfully strange.

Looking at less advanced stats on ESPN.com, Cabrera finished 6th of 14 qualifiers in fielding percentage and 7th of 14 in range factor. Seems pretty average. Also, for what it's worth, he was 2nd among MLB third basemen in turning double plays. This was admittedly helped by his playing time, but he turned far more than Headley or Wright, who played a similar amount of games.

No one's arguing that Cabrera's an excellent 3B, but a lot of us think he's legitimately "un-terrible."

For baserunning, Baseball-Reference has Cabrera at exactly "0" for Rbaser (Runs from Baserunning). I don't think that qualifies as terrible either.

GIDP, I'll grant, was pretty terrible.
   12. Bring Me the Head of Alfredo Griffin (Vlad) Posted: November 25, 2012 at 10:53 PM (#4309387)
Granted, no one wants to be last on any list, but this also shows that he was good enough to play every day at 3B.


I think that overstates it. At most, it shows that the Tigers didn't have any better options.

Looking at less advanced stats on ESPN.com, Cabrera finished 6th of 14 qualifiers in fielding percentage and 7th of 14 in range factor. Seems pretty average.


Neither of those statistics is a particularly good measure of defense.
   13. something like a train wreck Posted: November 25, 2012 at 11:12 PM (#4309400)
Neither of those statistics is a particularly good measure of defense.

What is the empirical validation that UZR is? Serious question, not a taunt. I understand the theory, but how do we test that it works?
   14. something like a train wreck Posted: November 25, 2012 at 11:14 PM (#4309401)
Neither of those statistics is a particularly good measure of defense.

What is the empirical validation that UZR is? Serious question, not a taunt. I understand the theory, but how do we test that it works?
   15. Cooper Nielson Posted: November 25, 2012 at 11:43 PM (#4309415)
Granted, no one wants to be last on any list, but this also shows that he was good enough to play every day at 3B.

I think that overstates it. At most, it shows that the Tigers didn't have any better options.


Glass half-full/half-empty. They didn't have better options because they didn't think they needed them. Cabrera started 154 games at 3B, 7 at DH, and I think most of the DH games were because he was hurt. He was rarely removed for a defensive replacement. They released Brandon Inge in April, then didn't even choose to carry a true 3B on the bench. And after this year's "experiment," there's no question that Cabrera will be back at 3B next year. The Tigers don't seem to see his defense as a big problem, either because they're stupid, or because it wasn't. I lean toward the latter, though others might disagree.

Looking at less advanced stats on ESPN.com, Cabrera finished 6th of 14 qualifiers in fielding percentage and 7th of 14 in range factor. Seems pretty average.

Neither of those statistics is a particularly good measure of defense.


Of course they're not, that's why I called them "less advanced." However, the subject at hand is whether Miguel Cabrera is a "terrible" third baseman, or a "passable" third baseman.

We all have our own definitions of "terrible," but to me a "terrible" third baseman would make a lot of errors and would look bad out there. Cabrera did not make a lot of errors and he looked pretty comfortable.

Even when you look at the most "accurate" advanced stats, he was maybe at the 20th-25th percentile of all third basemen last year in terms of capability (counting everyone who played 3B, not just the full-timers) -- certainly not the 5th or 10th percentile, which is where I would put "terrible."
   16. Bring Me the Head of Alfredo Griffin (Vlad) Posted: November 25, 2012 at 11:59 PM (#4309424)
What is the empirical validation that UZR is? Serious question, not a taunt. I understand the theory, but how do we test that it works?


Check the correlation for team UZR against runs allowed?

Glass half-full/half-empty. They didn't have better options because they didn't think they needed them.


Or they weren't able to acquire them, in spite of need.

We all have our own definitions of "terrible," but to me a "terrible" third baseman would make a lot of errors and would look bad out there.


Some terrible defenders hardly ever make errors because they don't have enough range to get their hands on balls. Think late-career Luke Easter.
   17. Cooper Nielson Posted: November 26, 2012 at 12:16 AM (#4309432)
Some terrible defenders hardly ever make errors because they don't have enough range to get their hands on balls. Think late-career Luke Easter.

Huh? Though I accept your general point, Luke Easter, playing 1B, made 14 errors (125 games), 18 errors (118 games), and 9 errors (56 games) in his last three years. That's a heckuva lot of errors for a first baseman.

Cabrera was 4th among MLB third basemen this year in total chances, 1st in putouts, 8th in assists, and 2nd in double plays. As acknowledged before, those are all correlated with playing time (and depend somewhat on pitching staff), but it doesn't suggest that he's a guy who just stands and watches balls go by. Old-school, semi-accurate "range factor" is based on chances per game (I think), and his range factor shows him as being pretty average. He's not Luke Easter.

For those of you who think Cabrera was a "terrible defender" this year, I want to humbly ask you: How much did you actually see him play? I know he has been very bad in the past, and the advanced stats are not particularly friendly to him, but if you watched him play in 2012, I find it hard to believe you would call him a "terrible" third baseman.
   18. PerroX Posted: November 26, 2012 at 12:48 AM (#4309456)
Malcolm makes an excellent point -- if the statheads hadn't been quite so dogmatic, their arguments would have played so much better. Try being reasonable next time and not like Romney supporters on 11072012.
   19. Bring Me the Head of Alfredo Griffin (Vlad) Posted: November 26, 2012 at 01:03 AM (#4309464)
Huh? Though I accept your general point, Luke Easter, playing 1B, made 14 errors (125 games), 18 errors (118 games), and 9 errors (56 games) in his last three years. That's a heckuva lot of errors for a first baseman.


I was talking about the end of Easter's pro career, not the end of his MLB career. He was out of the majors in 1954 (mostly because he couldn't run anymore), but active at AAA through 1963, and his immobility and inflexibility in those last couple of years cut down on his error totals considerably. They were still high, but not as high as they would've been if he could actually bend down and touch his shoes.
   20. Bring Me the Head of Alfredo Griffin (Vlad) Posted: November 26, 2012 at 01:04 AM (#4309465)
Malcolm makes an excellent point -- if the statheads hadn't been quite so dogmatic, their arguments would have played so much better.


Why should the people who are right on the facts be forced to compromise with the people who aren't? That's the kind of thinking that makes legislators round Pi down to an even 3.
   21. Booey Posted: November 26, 2012 at 01:08 AM (#4309469)
Malcolm makes an excellent point -- if the statheads hadn't been quite so dogmatic, their arguments would have played so much better


I've kinda wondered if this kind of thing happened with WAR too; I doubt people would have had too much trouble believing that Trout may have actually been better than Cabrera, but it seemed like it was a bit unbelievable to lots of people that the difference could possibly be THAT great (or that Trout's year wasn't just the best this season but also one of the best of all time). It seems the enormous WAR gap may have turned some people off from the credibility of the stat altogether. I actually wonder if the vote would have been slightly closer if their WAR was a bit closer.
   22. PerroX Posted: November 26, 2012 at 01:14 AM (#4309473)
Why should the people who are right on the facts be forced to compromise with the people who aren't? That's the kind of thinking that makes legislators round Pi down to an even 3.


It's the pretension that WAR is the same kind of thing as Pi that kills your argument.

In much smarter places than these threads, I've seen folks intelligently argue for for WAR in the overal context of an MVP discussion, and I can agree that Trout over Cabrera is a sound and reasonable opinion. But pretty much all I've read here are people knocking down straw men (thanks Repoz, for setting them up) and making harebrained arguments like you should add four home runs to Trout for his defense.

I know more than a couple of you have a solid grasp of WAR, but the discussions here rarely demonstrate it. Instead, as usual, it's a race to see who can proclaim, "I'm right, you're wrong" the most times. And Jim basically led the charge of the right brigade.

If any of you want to present a reasoned argument for Trout, make it again. Cut and paste if you wish.
   23. Bring Me the Head of Alfredo Griffin (Vlad) Posted: November 26, 2012 at 01:16 AM (#4309478)
If any of you want to present a reasoned argument for Trout, make it again.


He's the better player, and his performance was worth more runs than Cabrera's.
   24. Honkie Kong Posted: November 26, 2012 at 01:21 AM (#4309485)
Why should the people who are right on the facts be forced to compromise with the people who aren't?

Except in this case, you are not right. The bulk of the difference in their value comes from defence, which the creators of the stat themselves claim has little substance as a single season metric, and has to be heavily regressed to a 3 year average.
Given that, the difference in their value is not that large, and there in lies the point which I think Don Malcolm is making.
Trout was better that season still, but it is not the current mindnumbing gulf between what the metrics are saying and what the observer is perceiving.
   25. tshipman Posted: November 26, 2012 at 01:23 AM (#4309487)
I've kinda wondered if this kind of thing happened with WAR too; I doubt people would have had too much trouble believing that Trout may have actually been better than Cabrera, but it seemed like it was a bit unbelievable to lots of people that the difference could possibly be THAT great (or that Trout's year wasn't just the best this season but also one of the best of all time). It seems the enormous WAR gap may have turned some people off from the credibility of the stat altogether. I actually wonder if the vote would have been slightly closer if their WAR was a bit closer.


This is a lot of it for me. TBH, the loud harping after the fact made me a lot more uncertain about the methodology.

Angel Stadium, after being an average park for years, is Petco II?
Batting leadoff can make up for missing 30 games?
Part-time LF's barely see a ding on their positional adjustment?
   26. PerroX Posted: November 26, 2012 at 01:23 AM (#4309488)
I like the concept of WAR, but it's not as cut and dried as something like OPS+. My chief objection has been that Trout didn't play the first month of the season and was outperformed the last two months at the plate by a sizeable margin. WAR would argue that May through July was so overwhelmingly valuable as to override the season as a whole, then adding in baserunning and defense, no contest. I'm willing to say maybe. But you should think about arguing the point persuasively with the guys who play and cover the game and not just sneer at them from your mother's basements.

Sabermetrics has made its headway by doing just that.
   27. Cooper Nielson Posted: November 26, 2012 at 01:26 AM (#4309490)
I was talking about the end of Easter's pro career, not the end of his MLB career. He was out of the majors in 1954 (mostly because he couldn't run anymore), but active at AAA through 1963

Well, now you're just getting obscure. :-) Sorry I am not familiar with Luke Easter's minor league career in his 40s (before I was born).

Though, once again, looking at BB-Ref's stats, Easter made plenty of errors. From 1955 to 1963 (ages 39-47) -- again, all as a first baseman -- he made 30(!), 12, 14, 20, 20, 7 (78 games), 2 (53 games), 16 (67 games), and 6 errors (47 games).

In 1961 (2 errors in 53 games) I see your point about "Some terrible defenders hardly ever make errors." In the rest of the years, his 1B error rates range from high to alarming.
   28. PerroX Posted: November 26, 2012 at 01:26 AM (#4309491)
Obviously there's a lot of intelligence at work in these threads. So apologies if I painted with too broad a brush.
   29. PerroX Posted: November 26, 2012 at 01:34 AM (#4309498)
He's the better player, and his performance was worth more runs than Cabrera's.


I agree he's the better overall player, even just judging last year's performance. But I think we'd benefit from taking a broader historical outlook at the award, especially within the definition presented to the voters. Hitting is given the greatest attention, playing time considerable weight, and making the playoffs a final arbiter. Also, Cabrera's value as a hitter over his career added an intangible to his case.

The biggest question I'll ponder in reflection, after the dust settles, is what would have happened if Trout doesn't spend a month by The Great Salt Lake.
   30. bjhanke Posted: November 26, 2012 at 01:47 AM (#4309505)
Avoid running (comment #4) is perceptive. I've known Don for quite a while, and I don't think he ever wrote music criticism for anything larger than the Washington University student newspaper. But he knows a LOT about music. When he was living here in STL, I remember him getting a job at a music chain store called "Peaches" as their classical music department head or classical music buyer, something like that, involving classical music. That's indicative because Don's PRIMARY tastes in music are not for classical, but his knowledge of music is wide and deep. He even managed to get me to understand part of why Brian Wilson's Smile work is so good, and I'm a bit tone deaf. Teaching music to me is not easy. I am also completely unqualified to say whether his writing style is like that of music critics, because I almost never read any music criticism. I can't understand it. But the idea that music criticism would seep into Don's writing style is completely plausible. - Brock Hanke
   31. Bring Me the Head of Alfredo Griffin (Vlad) Posted: November 26, 2012 at 01:47 AM (#4309506)
Hitting is given the greatest attention, playing time considerable weight, and making the playoffs a final arbiter.


Yes, but it shouldn't be. That interpretation is contrary to the written voting guidelines (except for the part about playing time), and voters who vote based on those things are voting irrationally.
   32. steagles Posted: November 26, 2012 at 02:35 AM (#4309525)
What is the empirical validation that UZR is? Serious question, not a taunt. I understand the theory, but how do we test that it works?
Check the correlation for team UZR against runs allowed?
isn't that kind of the same thing as using the word you're defining in the definition of the word you're defining?

also, my problem with UZR is that it's not a descriptive statistic. if you look at fielding percentage, that tells you something that actually relates to the game as it is played, whereas UZR is just some amalgam of abstract concepts.



also, if miguel cabrera didn't have the triple crown going for him, if he had say, 5 fewer hits, and if josh hamilton had hit 2 more HRs, and he wasn't the leader in either of those categories, does anyone think he still wins the MVP? i guess it's possible, but it seemed as if the discussion for AL MVP ended with "triple crown" for a lot of people, and they didn't consider any other argument.

also, w/r/t this:
Malcolm makes an excellent point -- if the statheads hadn't been quite so dogmatic, their arguments would have played so much better. Try being reasonable next time and not like Romney supporters on 11072012.
could really be used to our advantage if we just harnessed its potential. are you a stathead who wants ichiro in the hall of fame? then all you have to do is talk about how overrated hits are, and about how adam dunn was a more valuable hitter because all of his walks!!!

   33. valuearbitrageur Posted: November 26, 2012 at 03:04 AM (#4309540)
No one's arguing that Cabrera's an excellent 3B, but a lot of us think he's legitimately "un-terrible."


No one's arguing he's a good defensive 3B either. His career record at 3B corresponds to being a bad defender (-6.4 runs/150), by UZR and other measures. Watching him this year I saw a surprisingly agile athlete for a man of his size with soft hands and the range of a statue, which is pretty much how UZR measures him. And if you lower the innings limit on UZR to 600 from the 900ish it uses as "qualified", Cabrera is 28/31 in UZR rate, and 30/31 in total UZR value.

Now maybe UZR is off this year due to sample size and he's actually only -6 runs/150 instead of -11/150 it thought he was this year. But even that's a pretty big stretch, by age 25 he had almost 2.5 full years averaging -4/150, since he came to the Tigers he's got one full year averaging around a -14/150, mostly in this, his age 29 year. It's hard to imagine he's still "only" -4 or -6, both of which are still bad.

But again, my original comment wasn't that he was terrible at defense, I said he was terrible at everything that wasn't slo pitch softball stats (OBP/hitting for average/ power and driving in the multitude of extra base-runners he was given over Trout).

Even if you think Cabrera was just a bottom quartile defender, that makes him a poor defender.

He was also a bottom quartile base-runner. He's 16/18 third basemen in base-running.

He's #1 in the league in GIDP.

Cabrera was -12.9 runs in UZR & Base Running. No third baseman is close, except for Chris Johnson, and Cabrera grounded into 10 more double plays than Chris Johnson. Even if you discount UZR, Cabrera was clearly the worst third baseman in the league at everything other than hitting for average and power. In total, he's the best 3B in the league at giving away outs when he's not hitting. That's not terrible?

If you compare him to everyone, not just third basemen, Cabrera is 4th out of 143 qualified players in giving away the most runs in UZR and Base Running, the only players "better" than him in that measure are Jeter, Granderson, and Sin-Soo Choo. The only one close to Cabrera in GIDP is Jeter. So of 143 qualified players, the two worst at everything that's not pure hitting are Jeter (-14.3 runs UZR/Baserunning & 24 GIDP) and Cabrera (-12.8 runs UZR/28 GIDP). Whether Jeter wins, or Cabrera does is immaterial, because second worst out of 143 players is clearly terrible.

Even if you quibble with UZR and other stats, and think he's only 130th or 120th, well, that is still very bad. If you won't admit it's awful, you have to admit it ain't passable, that's for sure.
   34. Bring Me the Head of Alfredo Griffin (Vlad) Posted: November 26, 2012 at 03:15 AM (#4309547)
if you look at fielding percentage, that tells you something that actually relates to the game as it is played, whereas UZR is just some amalgam of abstract concepts.


How do you feel about OPS+?
   35. steagles Posted: November 26, 2012 at 03:37 AM (#4309553)
How do you feel about OPS+?
i'm fine with it. if you asked me to break OPS+ down to its components, i'd say it's OPS x 100 / league(slash)park adjusted OPS.

now, what would you say if i asked you to break down UZR in the same way?

   36. Bring Me the Head of Alfredo Griffin (Vlad) Posted: November 26, 2012 at 03:48 AM (#4309556)
i'm fine with it. if you asked me to break OPS+ down to its components, i'd say it's OPS x 100 / league(slash)park adjusted OPS.


And what does raw OPS tell you about "the game as it is actually played"?

now, what would you say if i asked you to break down UZR in the same way?


I'd say that it's the number of runs saved (or scored) as a result of the fielder handling (or not handling) balls hit in his general area.
   37. Cooper Nielson Posted: November 26, 2012 at 03:53 AM (#4309557)
But again, my original comment wasn't that he was terrible at defense, I said he was terrible at everything that wasn't slo pitch softball stats (OBP/hitting for average/ power and driving in the multitude of extra base-runners he was given over Trout).

See, to me, this sounds even more damning than saying he was "just" terrible at defense. Really, you think Miguel Cabrera is terrible -- not just below average, but terrible -- at everything in baseball besides "slo pitch softball stats" (which, comically, covers every single aspect of "hitting")? You don't think he can throw? You don't think he has "baseball IQ"?

Even if you think Cabrera was just a bottom quartile defender, that makes him a poor defender.

Not necessarily, depends on the distribution. Would you say a 25th percentile finisher in Miss Universe is "ugly"? Would you say a 25th percentile student in an MIT math class is "stupid"? There are some very good defensive players at 3B. There are not too many bad ones, because most bad defensive players are not allowed to play 3B. The worst defensive everyday 3B is not necessarily terrible, and is not necessarily even bad. Our definitions apparently differ, but I'd say Cabrera's defense this year was closer to "passable" than "terrible."

Pretty much all of Cabrera's stats you cite as "terrible" are counting stats. Cabrera plays 160 games a year, of course he's going to be "ahead" in such stats in any areas where he's below average. (For example, if his "terrible" baserunning costs the Tigers 0.1 runs per game, he's going to generate more of said negative stat than a truly terrible guy who is -0.3 runs per game but plays 50 games. Same with his defensive stats when you compare him to occasional 3B like Michael Young or Mark Reynolds.)

As I mentioned, Baseball-Reference has him at "0" for baserunning -- perfectly average. This squares with my observation. He's not fast, but he makes pretty good decisions and hustles when necessary. This might or might not rank him 16th out of 18th among 3B (I am not sure what stat you are citing) but it is decidedly unterrible.

If you compare him to everyone, not just third basemen, Cabrera is 4th out of 143 qualified players in giving away the most runs in UZR and Base Running, the only players "better" than him in that measure are Jeter, Granderson, and Sin-Soo Choo.

So you're saying the four most terrible players in baseball (except in hitting) are Derek Jeter, Curtis Granderson, Shin-Soo Choo and Miguel Cabrera? I guess I should have just left Robert in Manhattan Beach's original comment alone:

And another person who doesn't know what the word 'terrible' means.
   38. valuearbitrageur Posted: November 26, 2012 at 04:33 AM (#4309562)
Not necessarily, depends on the distribution. Would you say a 25th percentile finisher in Miss Universe is "ugly"?


A bottom 25 percentile finisher? If it got a top 25% finisher to blow me, yes.

Would you say a 25th percentile student in an MIT math class is "stupid"? There are some very good defensive players at 3B. There are not too many bad ones, because most bad defensive players are not allowed to play 3B. The worst defensive everyday 3B is not necessarily terrible, and is not necessarily even bad. Our definitions apparently differ, but I'd say Cabrera's defense this year was closer to "passable" than "terrible."


I'm sure there are some terrible baserunners in the minors, in college ball and in high school baseball and in womans softball. Do you want to include them so you can make the case that Miggy isn't terrible because some (but not all) of female college softball players are even slower than him? There are 143 qualified players by Fangraphs, that's a pretty big pool to be in the bottom 4.

Pretty much all of Cabrera's stats you cite as "terrible" are counting stats. Cabrera plays 160 games a year, of course he's going to be "ahead" in such stats in any areas where he's below average. (For example, if his "terrible" baserunning costs the Tigers 0.1 runs per game, he's going to generate more of said negative stat than a truly terrible guy who is -0.3 runs per game but plays 50 games. Same with his defensive stats when you compare him to occasional 3B like Michael Young or Mark Reynolds.)


Since coming to the Tigers, Cabrera has played over a year at third with worse UZR than Michael Young or Mark Reynold's career averages. HIs career average isn't greatly better than theirs (-6 vs. -9 vs -11). All you are establishing is that when third basemen field third as badly as Cabrera has in his career, esp. recently, they are moved off third as quickly as possible.

As I mentionedBaseball-Reference has him at "0" for baserunning -- perfectly averageThis squares with my observationHe's not fast, but he makes pretty good decisions and hustles when necessary. This might or might not rank him 16th out of 18th among 3B (I am not sure what stat you are citing) but it is decidedly unterrible. 


I was citing FanGraph's numbers and again, I'm saying his aggregate value at all of these skills, defense/baserunning/hitting into double plays, was near the lowest in baseball. FanGraph's estimates his defense and base-running as being worth -12.8 runs this year, while Baseball Reference thinks Miggy was -9 runs total in running/defense/hitting into double plays. It looks like Fangraphs includes double plays in BsR, my mistake for calling it out separately, but it doesn't change his Fangraphs ranking much, if anything.

If you think BBRef doesn't rank him so badly in those categories, go ahead and show it. I don't have access to do those kind of reports on BBRef, which is why I used Fangraphs. But I doubt he finishes out of the bottom 10.

So you're saying the four most terrible players in baseball (except in hitting) are Derek Jeter, Curtis Granderson, Shin-Soo Choo and Miguel Cabrera? I guess I should have just left Robert in Manhattan Beach's original comment alone:


Not Granderson, UZR has him as a slightly plus defender during his career, he was a plus base-runner this year, and didn't hit into double plays much. His -17 UZR this year is far more likely to be temporary or a sample size issue than Cabrera's number is.

Probably not Shin-Soo Choo either. His career UZR was positive until this year, and even after this -17 year is only -2.7/150 in right for his career. He didn't ground into many DPs either (11) and his base-running is only slightly negative. So if his UZR sample size was a little fluky this year and he really was -12 or less, he wouldn't be as bad as Cabrera and would have 17 fewer GIDP.

Jeter was -15.2 in UZR. It wasn't his worst UZR year ever, his career UZR average is -6.5/150, but he turned 38 during the season so it's not unreasonable to think that his UZR score isn't that far from reality. He had positive base running value, but barely. And he hit into 24 double plays. So yea, it sure looks like Jeter was pretty terrible outside of BA/OBP/SLG this year.

Cabrera's defense wasn't far from his career averages and not far from what you'd expect from the 29 year old version of Miggy. He led the league in GiDP, and he was one of the worst in baserunning.

If the players who give up the most runs on defense, in base running, and in GiDP aren't terrible, what the hell does it take to be terrible?
   39. Cooper Nielson Posted: November 26, 2012 at 04:34 AM (#4309563)
Even if you quibble with UZR and other stats

UZR says that 2012's two Gold Glove winning centerfielders, Andrew McCutchen and Adam Jones, each cost their teams more outs on defense than Delmon Young* and Prince Fielder did. Does that make them "terrible" fielders? Because I'm pretty sure they're better defensive players than Delmon Young or Prince Fielder.

* I know, Delmon had limited playing time in the OF, but that's part of my point. The "terrible" defensive players move to DH or 1B, or don't play at all, thus they don't have a chance to rack up high negative UZR. And comparing total UZR across positions (including DH) seems impractical.
   40. steagles Posted: November 26, 2012 at 04:40 AM (#4309564)
I'd say that it's the number of runs saved (or scored) as a result of the fielder handling (or not handling) balls hit in his general area.


but how is it calculated? i mean, you described what UZR is supposed to mean, but what's the methodology that gets you there?


i'm a bit off when it comes to stats, because i tend to prefer specific descriptive statistics (since we're talking about fielding, i'd point to RZR and OOZ as far as stats that i like to use) as opposed to blackbox statistics that i'm just supposed to take at face value without really questioning how the statistic is created.
And what does raw OPS tell you about "the game as it is actually played"?
specifically? it's on base plus slugging. it's an approximation for offensive value based on hits, walks and total bases. it's not a perfect statistic (take for example the mismatched denominators of OBP (plate appearances) and slugging percentage (at bats)), but i've used it for so long that i just inherently understand what it's saying, and it's become a shorthand for discussions.

   41. steagles Posted: November 26, 2012 at 04:43 AM (#4309565)
and post 38 just broke the thread

   42. valuearbitrageur Posted: November 26, 2012 at 04:54 AM (#4309568)
It won't let me fix it. Honestly.
   43. Cooper Nielson Posted: November 26, 2012 at 05:03 AM (#4309569)
If you think BBRef doesn't rank him so badly in those categories, go ahead and show it. I don't have access to do those kind of reports on BBRef, which is why I used Fangraphs. But I doubt he finishes out of the bottom 10.

Well, I already said twice that BBRef has Cabrera at 0 Runs from Baserunning, which includes SB/CS, and I believe also includes all bases advanced for other events. (I am not 100% sure.) With his "0," he's in a huge tie for 156th out of 1284 players. I don't know how to limit it to "qualified" players but by definition a 0 is "average," not "among the worst." And while it's not particularly meaningful, he did steal 4 bases in 5 attempts.

Since coming to the Tigers, Cabrera has played over a year at third with worse UZR than Michael Young or Mark Reynold's career averages. HIs career average isn't greatly better than theirs (-6 vs. -9 vs -11). All you are establishing is that when third basemen field third as badly as Cabrera has in his career, esp. recently, they are moved off third as quickly as possible.

Now I am a little confused. Are we talking about career averages or 2012? Because I'm trying to specifically talk about his performance this year. I know he has been a lousy defender in the past. But it's possible to improve at defense if you put effort into it. I believe Cabrera did that. Maybe it's just a fluke, who knows? But in 2012, his defense at 3B certainly seemed "passable" from an observational standpoint, and even the stats that paint him in the worst light in 2012 do not seem overwhelmingly awful.

But maybe we simply have a different understanding of "terrible."

He seems to me to be:

* An average baserunner
* A below average (but not remarkably so) fielder at a difficult position
* Very bad at avoiding GIDPs

You can add up your numbers to show me that he and Curtis Granderson and Shin-Soo Choo were more "terrible at everything not related to hitting" than Delmon Young and Prince Fielder, but I will not believe you. (And shouldn't GIDP be included in hitting, not baserunning?)
   44. Cooper Nielson Posted: November 26, 2012 at 05:31 AM (#4309571)
If you think BBRef doesn't rank him so badly in those categories, go ahead and show it.

OK, for BBRef fielding (caveat: I'm not sure what all these numbers mean), here are the guys who played at least 50 games at 3B and had a worse "Total Zone Fielding Runs Above Average per 1,200 Innings" than Cabrera (-9) in 2012:

-9 Hanley Ramirez
-10 Jose Lopez
-10 Ian Stewart
-11 Alex Rodriguez
-12 Wilson Betemit
-14 Trevor Plouffe
-24 Jordan Pacheco
-25 Chris Nelson

And here are the guys with 50+ games who do worse than Cabrera's "BIS Defensive Runs Saved Above Average per 1,200 Innings" (Cabrera is at -4):

-5 Eric Chavez
-5 Pedro Alvarez
-6 Will Middlebrooks
-7 Kyle Seager
-7 Pablo Sandoval
-12 Chris Johnson
-12 Wilson Betemit
-12 Trevor Plouffe
-15 Hanley Ramirez
-24 Jordan Pacheco
-33 Chris Nelson

Here's BIS +/- Fielding Runs Above Average, which is a counting stat -- his playing time hurts him here -- where Cabrera was -8:

-9 Trevor Plouffe
-10 Chris Johnson
-10 Greg Dobbs (in only 36 games)
-13 Hanley Ramirez
-14 Jordan Pacheco
-17 Chris Nelson

(Note: Ryan Zimmerman was pretty close to Cabrera, at -7 in 145 games.)

Cabrera doesn't finish last in any of these unless you move the qualifying level up to 100 games. And, in fact, the distance between him and the worst guy is quite large in all cases. Now, if you're looking for a "terrible" third baseman (by the numbers), I'd look at Pacheco/Nelson, Hanley Ramirez, Trevor Plouffe, or Greg Dobbs. Cabrera seems clearly superior to all these guys if you believe these numbers.
   45. Robert in Manhattan Beach Posted: November 26, 2012 at 06:05 AM (#4309572)

But maybe we simply have a different understanding of "terrible."


Yes, you have the correct understanding, KT has the nonsensical one, which is what I pointed out in #3. I hate the 'we have different definitions' cop out. The word has one definition and KT is using it wrong.

Also, 'passable' doesn't mean average. It means not terrible.
   46. Cooper Nielson Posted: November 26, 2012 at 06:51 AM (#4309574)
I hate the 'we have different definitions' cop out. The word has one definition and KT is using it wrong.

Well, I was being gently sarcastic. :)
   47. RMc is a fine piece of cheese Posted: November 26, 2012 at 09:06 AM (#4309589)
Well, somebody broke the thread. Can't read anymore.
   48. fra paolo Posted: November 26, 2012 at 10:14 AM (#4309598)
It's hard to imagine [Cabrera]'s still "only" -4 or -6, both of which are still bad.

This is the essence of the argument about Cabrera's defence. It breaks down to three elements:

UZR, one of several measures of fielding introduced during the 21st century, rates Cabrera below average.
Cabrera's reputation is not as a glove man.
Therefore, Cabrera is a bad or terrible or some other negative adjective fielder.

However, these statements, concerning lists of the fourteen AL third-basemen with the most innings at the position are facts:

Cabrera scores well in Win Shares Above Bench's fielding, among AL third-basemen. This seems to be on account of his ability to be involved in double plays.
Cabrera's 2012 season at third base is not historically bad. Somebody has to be at the bottom of the list, and this year it is Cabrera.
Although Cabrera is at the bottom of all lists of UZR, BPro FRAA, Humphrys' DRA and Dewan's DRS, what makes interesting reading is that Adrian Beltre is at the bottom of the DRA and FRAA lists, too.

It is the third statement that is most salient here. Beltre is traditionally accounted as one of the best fielding third basemen in the league. Go check the HoM's MMP thread for 2012 to see this theme at work. Yet it does seem as if his ranking depends more on how one counts the numbers. And the same is true of Win Shares. Win Shares counts the numbers differently to the others.

The consensus of defensive measures is that Cabrera is a poor defender relative to other AL third basemen. But is a poor-fielding 3B still inherently more valuable than an excellent CF? Win Shares Above Bench suggests that this is the case. Trout has 1.1 Fielding WSAB, and Cabrera has 2.8. And this is the problem that proponents of UZR and other metrics fail to address. They emphasise the value of player relative to his peers at his position.
   49. Yeaarrgghhhh Posted: November 26, 2012 at 10:24 AM (#4309599)
Maybe I just wasn't paying attention, but what are some examples of stat people being incredibly dogmatic about WAR and Trout during the season? I saw people cite it as evidence, but I don't recall many people arguing that Trout was unquestionably the better player because WAR said so. If anything, it seems like WAR has been used more dogmatically by anti-stats folks as a bludgeon.

I also think the argument that WAR isn't helpful in this debate because the defensive component isn't hard evidence to be a red herring. We don't need UZR or WAR or any other advanced stat to see that Trout is an above average CF and that Cabrera is an ok 3b. So Trout was a better fielder at a more important position. Is there really any debate about that? Given that Trout was also one of the best baserunners in baseball, almost as good a hitter as Cabrera, and his team was just as good, the MVP argument is really very simple.
   50. fra paolo Posted: November 26, 2012 at 10:29 AM (#4309601)
So Trout was a better fielder at a more important position. Is there really any debate about that?

Yes. It may be that 3b is the more important position. Tango's positional adjustments put them exactly level, at +2.5 runs.
   51. Yeaarrgghhhh Posted: November 26, 2012 at 10:38 AM (#4309603)
Fine, but even if you assume the positions have equal value, it's quite a stretch to argue that Cabrera provided as much value defensively. I'm not up in arms about Cabrera winning, and I can see the argumetn that maybe Cabrera is only a -5 3B and Trout is only a +5 CF, so the gap isn't *that* large, but the notion that defense is basically a wash doesn't pass the smell test.
   52. Bring Me the Head of Alfredo Griffin (Vlad) Posted: November 26, 2012 at 10:45 AM (#4309608)
but how is it calculated? i mean, you described what UZR is supposed to mean, but what's the methodology that gets you there?


A trained observer watches every ball hit in the fielder's general area, and logs whether or not he made the play, and in which defensive zone the play took place. Then, at the end of the game, they add up the run value of each play in a fielder's zone of responsibility, add it to his total, and show the result when compared to a league baseline. That's how I understand it, anyway - apologies if that's incorrect.

Is there a certain amount of discretion involved? Sure. But there's that same discretion that's used in more traditional fielding stats, when the official scorer determines whether a given ball is a hit or an error.
   53. fra paolo Posted: November 26, 2012 at 11:05 AM (#4309622)
it's quite a stretch to argue that Cabrera provided as much value defensively.

Using .26 for the out and .44 for the error.

Cabrera = 127 put outs + 243 assists - 13 errors = 90.5 runs.
Trout = 340 put outs + 3 assists - 4 errors = 87.4 runs.

Now, I've been around long enough to know that there is plenty to quibble with here, and have done some quibbling myself.

But I don't think it's at all a stretch to argue that Cabrera provided as much value defensively. Win Shares Above Bench does so, and then some. It depends on how one wants to allocate value among fielders.

The fact is, a team needs to provide 162*27 outs = 1,268 runs of defensive value (again using .26 for the out) every season, give or take extra innings and not scoring enough runs to take the game into the bottom of the ninth, neither of which is necessarily the fault of the fielders as fielders. There's an absolute value as well as a relative one to be taken into account.
   54. Jittery McFrog Posted: November 26, 2012 at 01:42 PM (#4309776)
if you asked me to break OPS+ down to its components, i'd say it's OPS x 100 / league(slash)park adjusted OPS.


That's not correct.

OPS+ is calculated that way because it makes the coefficients come out sort of close to their
linear weights values. It's really a Frankenstein stat, and not nearly as straightforwardly
helpful as many people think.
   55. Walt Davis Posted: November 26, 2012 at 03:58 PM (#4309889)
This is a lot of it for me. TBH, the loud harping after the fact made me a lot more uncertain about the methodology.

Angel Stadium, after being an average park for years, is Petco II?


Angels home/road
Opps Anaheim/home

2012:
272/325/428, 82 HR, 348 R vs. 276/337/437, 105 HR, 419 R
234/291/367, 79 HR, 309 R vs. 259/329/440, 107 HR, 390 R

2011:
248/310/377, 62 HR, 305 R vs. 257/315/426, 93 HR, 362 R
244/303/371, 69 HR, 297 R vs. 259/322/405, 73 HR, 346 R

2010:
247/313/378, 69 HR, 318 R vs. 250/309/402, 86 HR, 362 R
243/313/369, 68 HR, 322 R vs. 270/343/440, 80 HR, 380 R

Some totals:

At Anaheim: 429 HR, 1899 runs
Away: 544 HR, 2259 runs

H/R:
79% HR, 84% runs

There's nothing mysterious in park factors. There's no WAR magic in there. Just plain basic facts. Doesn't really matter why Anaheim is such a tough place to hit now, it is.

And it's interesting that nobody complained about the "unbelievability" of Anaheim's park factor until they needed an excuse to vote Cabrera.

And it's interesting that:

Trout home: 318/390/586 976 OPS, 16 HR
Trout road: 332/407/544 951 OPS, 14 HR

Cabrera home: 332/403/692 1094 OPS, 28 HR
Cabrera road: 327/384/529 913 OPS, 18 HR

But somehow it's Trout who's unfairly advantaged by his home park.
   56. Cooper Nielson Posted: November 26, 2012 at 10:19 PM (#4310163)
And it's interesting that nobody complained about the "unbelievability" of Anaheim's park factor until they needed an excuse to vote Cabrera.

To be fair, no one really cared if Mark Trumbo's OPS+ was "accurate." Park factors aren't complained about unless people notice them, and a lot of us didn't realize Anaheim was now a pitchers' haven until Mike Trout put up 10 WAR.

And it's interesting that:

Trout home: 318/390/586 976 OPS, 16 HR
Trout road: 332/407/544 951 OPS, 14 HR


See, that doesn't look like an extreme pitchers' park! :-)

You must be Registered and Logged In to post comments.

 

 

<< Back to main

BBTF Partner

Support BBTF

donate

Thanks to
cardsfanboy
for his generous support.

Bookmarks

You must be logged in to view your Bookmarks.

Hot Topics

Newsblog2014 WORLD SERIES GAME 2 OMNICHATTER
(526 - 11:46pm, Oct 22)
Last: boteman

NewsblogHow Wall Street Strangled the Life out of Sabermetrics | VICE Sports
(9 - 11:39pm, Oct 22)
Last: James Kannengieser

NewsblogOT: Politics, October 2014: Sunshine, Baseball, and Etch A Sketch: How Politicians Use Analogies
(3191 - 11:32pm, Oct 22)
Last: The Yankee Clapper

NewsblogMcSweeneys: NEW BASEBALL STATISTICS.
(9 - 11:25pm, Oct 22)
Last: JE (Jason)

NewsblogStatcast: Posey out at the plate
(14 - 11:25pm, Oct 22)
Last: cardsfanboy

NewsblogOT: Monthly NBA Thread - October 2014
(343 - 11:01pm, Oct 22)
Last: madvillain

NewsblogRoyals are not the future of baseball | FOX Sports
(29 - 11:00pm, Oct 22)
Last: boteman

NewsblogSielski: A friend fights for ex-Phillie Dick Allen's Hall of Fame induction
(181 - 10:59pm, Oct 22)
Last: theboyqueen

NewsblogCardinals proud of fourth straight NLCS appearance | cardinals.com
(65 - 10:38pm, Oct 22)
Last: cardsfanboy

NewsblogJerome Williams re-signs with Phils
(9 - 10:21pm, Oct 22)
Last: stevegamer

NewsblogJay set for surgery — and for CF in 2015 : Sports
(5 - 9:58pm, Oct 22)
Last: cardsfanboy

NewsblogAd Week: What Is Madeleine Albright Doing on the Wheaties Box?
(1 - 9:54pm, Oct 22)
Last: Howie Menckel

NewsblogHunter Pence responds to Royals fan signs with monster Game 1 | MLB.com
(54 - 7:50pm, Oct 22)
Last: JE (Jason)

NewsblogMike Scioscia, Matt Williams voted top managers
(43 - 7:45pm, Oct 22)
Last: catomi01

NewsblogDealing or dueling – what’s a manager to do? | MGL on Baseball
(39 - 7:10pm, Oct 22)
Last: bobm

Page rendered in 0.8374 seconds
52 querie(s) executed