Evans Clinchy. That can't be a real name.

Since I was expecting to be surprised, I though the article was going to tell how Jacoby was good defensively this year, after all.

they also have mulled moving him to left field

Please don't mull anymore, about this. You aren't expected to throw out runners from CF, but you have to, occasionally, from left, at Fenway. He won't.

I thought Ellsbury's defense was so good, he was actually a better overall player than Grady Sizemore.

To anyone who watched the Red Sox closely this season, that statement sounds like blasphemy. Ellsbury captivated Sox fans all year long with his exceptional range and dazzling highlight-reel catches in center field. If anything, we thought, Ellsbury was one of the strong points of the Red Sox’ defense this season, not one of the weaker ones.

Really? Because most guys I know who saw alot of Sox games this year said he took bad lines to ball, played too deep and got late jumps on just about everything...hence deserving his fairly crappy UZR.
His speed presents a two-fold problem. One it hides all his misgivings as a fielder to the naked eye(and average Joe fan) and two, because he's fast they insist on putting him in the leadoff spot.His OBP skills just aren't good enough with better alternatives available. Pedroia should bat first and Drew second and Ellsbury should be either 8th or 9th(but hey, that's my rant)

Defensive ratings seem all over the map. I can't imagine anyone who reads them with regularity being surprised by anything.

I thought Ellsbury's defense was so good, he was actually a better overall player than Grady Sizemore.

That's because you spent too much time listening to Kevin.

Hasn't MGL said time and time again not to take single year UZR's too seriously? Ellsbury was 7th among 63 MLB OFers last year. Granted, half of that was playing the corners, but I think it's fair to conclude that the jury is still out on him.

Evans Clinchy. That can't be a real name.

"Hey, you don't have to take that from no punk-assed crab!"

Hasn't MGL said time and time again not to take single year UZR's too seriously? Ellsbury was 7th among 63 MLB OFers last year. Granted, half of that was playing the corners, but I think it's fair to conclude that the jury is still out on him.

Exactly. Ellsbury's three-year UZR looks about average. I think he's a bit better than that, and I wouldn't be surprised if there's a park effect getting in there, but UZR seems to reflect reality pretty well as long as you don't try to slice it up beyond what the stat can stand.

Just how hard is it to learn to properly read fly balls? Is this kind of thing a vision problem or something, inability to properly visualize the flight of the ball as soon as you can? How do you get to be age 25, having played center for a goodly chunk of your life, and not learn these things? Not being snarky here and trying to defend Jake (in a roundabout way), I am genuinely curious. Is there often inherent ambiguity in the initial flight paths of some balls (the looping liner which at first appeared to be headed over your head, or vice-versa), and if so how does an outfielder learn to tell them apart? [This only occasional beer-league softball player doesn't quite have the experience required]

I thought Ellsbury's defense was so good, he was actually a better overall player than Grady Sizemore.

sure as hell had a better year.

to my eyes Ellsbury is good out there in the field, and Bay is a Canadian Manny

The Fenway park effect in CF is apparently that you seem insanely great one year then terribly awful the next. Over the past 4 years at least.

Grady Sizemore is still on the Cleveland Indians?

How do you get to be age 25, having played center for a goodly chunk of your life, and not learn these things?

Just conjecture, but I'd guess the speed helps cover any problems in reads/breaks.

I thought Ellsbury's defense was so good, he was actually a better overall player than Grady Sizemore.

Who are you talking to? Is he here? Then maybe you should let it go.

As for Ellsbury, yeah, he's athletic and makes some sensational plays, but he does take his share of bad routes - he also seems to do the "take a step back, oh #### it's shallower than I thought!" thing more often than opposing centerfielders (obviously a HUGE grain of salt here), and it's no secret his arm could charitably be described as "a little better than Johnny Damon", so it doesn't surprise me that he's not gold glove caliber by the metrics. And yeah, he was like +7/150 last year, in a smaller sample.

Looking ONLY at the numbers, it's probably too soon to say more than "he's almost certainly not elite right now, and he's probably been slightly below average overall to this point in his career." Anything beyond that, and it'd have to be based on some actual scouting analysis.

Is there often inherent ambiguity in the initial flight paths of some balls (the looping liner which at first appeared to be headed over your head, or vice-versa), and if so how does an outfielder learn to tell them apart?

Dan Okren't Nine Innings talks about this. Does anyone have that book handy? I may be misremembering things, but I think Gorman Thomas or Paul Molitor said that the atom ball was thoughest for centerfielders.

I may be misremembering things, but I think Gorman Thomas or Paul Molitor said that the atom ball was thoughest for centerfielders.

From my time in center (actually any outfield spot, but it's most common in center), that was certainly the case. If the ball is in any way angling away from you off the bat, it's easier to judge how hard it's been hit. But one hit directly at you is more difficult to determine how far it will travel. Obviously, major league center fielders should be much more adept at making a split-second determination, but I suspect it's still the toughest ball to judge for them as well.

Who are you talking to? Is he here? Then maybe you should let it go.

Seconded.

I was a much better CF than playing left or right. I didn't judge the ball hit right in my direction, but could run like the wind. In center field there are are more chances to run down balls hit to angles.

I've only got Ellsbury at -2, not as bad as his UZR. Defensive stats have enough noise as it is but there's also a ton of noise in the park factors used for Fenway. I've even got Bay a little above average this past year, but he was negative last year, in both Boston and Pittsburgh, and bad in 2007 as well, so the likely verdict on him is still that he's not a good fielder.

With 81 games in Fenway's LF, though, Ellsbury's arm doesn't come into play as much (as I would guess there aren't that many sends from 2nd base on balls hit to LF compared to normal parks), so he would perhaps be passable in LF.

I've argued elsewhere that perhaps the 1-2 year fix here is to move Ellsbury to LF and sign Cameron to play CF. The basis for this argument goes something like this:

1) Holliday isn't signing for less than top market price/years, and the Red Sox won't go there (correctly, in my view). The guess here is the Yankees sign him.
2) Bay may well sign on the west coast (he's instantly make the Giants a hell of a lot better and be their cleanup hitter), and for more $ than the Red Sox think he's worth.
3) There's really not any better options for LF unless you're pulling off some sort of trade.
4) Cameron may be willing to take a 2-3 year deal, which is all you really would prefer to commit if you can swing it.
5) Cameron upgrades the defense in CF and LF by moving Ellsbury over to LF.

Cameron obviously doesn't replace Bay's bat, but he's not wildly worse than Bay, and if you upgrade in other spots (SS, I'm looking at you) you can afford to lose a bit of production in the CF/LF slots. Add to this 6 months of Victor at C instead of 4 months of suck and 2 months of Victor and the offensive hit might not be as bad as it looks initially, while you upgrade the OF defense, perhaps by a whole lot.

Again, I'm not saying it's the BEST plan, all I'm saying is that in a scenario where we get neither Holliday nor Bay for LF, it may be the best plan B that's out there, barring some sort of trade (and who they'd trade for that would be worth it for LF/CF is not clear to me).

Ellsbury's fine cf. Two things. he doesn't crash land into fences and turf just to look like a Jeter showoff and plays the peculiar deep box Fenway cf which takes some range off the sides. UZR doesn't account for this stuff.

A question: Is there a way to figure out the Visiting Team's LF UZR for the 81 games at Fenway (the composite UZR)? I realize that doing so introduces a hell of a lot of noise to work with, since there could be upwards of 40 LFs in the mix, but if that composite UZR is about the same as Bay's LF UZR for the 81 games at Fenway, then doesn't that suggest that perhaps LF in Fenway really ############ the data to such an extent it's almost impossible to provide any clear picture?

My apologies if this has been answered in spades somewhere; I'm a bit of a novice as to the nuances between the various defensive metrics. I also don't really understand how the wall is adjusted for in the UZR method. I would guess once the computer system is in place, they'll be able to tell--precisely--how many balls hit off the wall higher than 10 feet versus ones that hit off the wall below that (10 feet would be the height of a wall you could reasonably expect a player to perhaps make an "over the wall" catch on--anything above that would probably be a HR in most parks, regardless of the effort of the player). That kind of data, it seems, would help sort out the wall factor a bit.

How much do park factors figure in with UZR? I've been wondering this for a while. All the A's right fielders are way positive by UZR, even guys like Swisher and Buck who don't look great, and since UZR's inception, almost every Athletic who has played RF has been twice as positive there as they have been in LF, including Cust. I've always figured it was an unaccounted for park factor.

Dan Okren't Nine Innings talks about this. Does anyone have that book handy?

I do, right next to me.

I may be misremembering things, but I think Gorman Thomas or Paul Molitor said that the atom ball was thoughest for centerfielders.

Let me look.

Okay, so I found what has to be what Daly's talking about here, pages 45-51 (if someone wants to check my work) are about Molitor taking over for Thomas in center at the beginning of 1981. A couple of quotes:

Thomas, on being moved to right field: "In center field, you're the driver in a grand prix race; in right, you're a mechanic."

"Thomas bore a month-long scowl on his face but worked hard at his new position, even facing the prospect of what he considered the hardest adjustment-having to field the line drive in the lights. 'People don't understand how scary it is,' Thomas told one who didn't understand. 'It's like looking into a flashbulb' - with a sphere hurtling at you at a hundred plus miles per hour, with the game on the line, with fifty thousand people watching."

"Suplizio worked on...and on his running, mostly on his running, despite Molitor's sizzle afoot: center fielders, Suplizio pointed out, need to run straight after the ball, shoulders square to the body, no turning of the head to follow the ball in flight. One day in Mesa, late that spring, the Brewers had just finished a game with the Cubs. Suplizio approached Molitor and said, 'I finally saw the back of your head today when you were chasing that gapper. You're a major league center fielder.'"

How much do park factors figure in with UZR?

Not at all, unless it changed when mgl went to work with the Cardinals. That's the whole reason why you have to take Manny's numbers with a huge graind of salt, as mgl has himself said. Now, that doesn't mean Manny's a true 0 defender, but it's a reason to take the -40s he was putting up and cut that back to say, -20.

Not at all, unless it changed when mgl went to work with the Cardinals. That's the whole reason why you have to take Manny's numbers with a huge graind of salt, as mgl has himself said. Now, that doesn't mean Manny's a true 0 defender, but it's a reason to take the -40s he was putting up and cut that back to say, -20.

I'm fairly certain that MGL has made a lot of park adjustments to his uzr in the last couple of seasons. Especially in regards to Fenway. Tango mentioned it on a thread sometime this week.

But the numbers don’t lie.

Well thats just bullcrap.
I'm ok with someone saying that about offense but even the best defensive metrics out there aren't close to being infallible.
I'm not saying the numbers are lying but you can't say defensive numbers can't be wrong and keep a straight face.

But the numbers don’t lie.

Defensive numbers lie through their teeth while farting "Dixie".

I watched Ellsbury a lot, and I'll second what a lot of others around here have said -- he's fast, he takes some funny routes, and my guess is that the whole package is somewhere around average. I think the idea of moving him to left, however, is folly. Running his bat out there every day, to the easiest field in all of baseball to play bar none, is a Waste with a capital W. I suspect that part of the reason the Sox have not much worried about LF defense is that it is their opinion -- as it is mine -- that LF defense, already not exactly a game-changer, is not terribly relevant in Fenway Park. He will never make up with his glove what he sacrifices with his bat. All of that for the privilege of betting that this isn't the year Cameron's legs die? Thanks, but no thanks.

Jeff K: you are mistaken. If you read MGL's article on UZR (introduced on Primer), you will see that Park Factors is prominent.

Jeff K: you are mistaken. If you read MGL's article on UZR (introduced on Primer), you will see that Park Factors is prominent.

Blah. I almost went back to reread it, then decided my memory was correct. I've read it, of course, I even linked it elsewhere on the site: Blogpark Learning Corner: UZR. Fat lot of good that did my memory.

I have no idea why I didn't remember, either, as I *do* remember the discussion about how the PFs are both calculated and applied from when those articles were first posted. My only defense is that they weren't presented until Part II, so I will channel kevin and defend myself on those grounds. :)

So tango, are Manny's -40s a mirage/overstatement because of the impossibility to truly adjust for Fenway LF? I've heard other defensive stat creators/defenders cite that impossibility, but I thought that had been overcome. Or is it something else? We've discussed those numbers so many times over the last 8 years that it's hard to find a specific instance of mgl talking about them, even though I know he's done so 10+ times at least.

MGL had a blog post at SOSH talking specifically about Manny.

In order to "prove" that the adjustment for LF/Fenway is fine, just add up the UZR for all fielders (Boston, opponents) at Fenway. You will get something close to zero.

Now, I don't think you have enough data publicly available to prove that, but you can certainly go through the team stats each year, and see how all non-Manny LF have done since 2002, and see if it looks right for those guys.

In my system, WOWY, the adjustment is pretty easy. Let me see what I have for Manny...

From 2001-08, he made 175 fewer plays in LF than other LF who played in the same parks he did. That's about 5.2 seasons, and that puts him at about 33 plays below average, or about 25 runs.

In RF (1993-2002), he as 91 fewer plays, in 5.6 seasons, or 16 plays below average, compared to other RF who played in the same parks he did.

That's all parks, weighted by his playing time at those parks.

I dunno... sounds pretty straightforward.

Manny's UZR (according to Fangraphs, which uses BIS data), 2002-2009, is -12 runs per season. So, that's pretty tame compared to what I have him at.