Does The Green Monster Really Skew Fielding Statistics?
[EDITOR'S NOTE: I wrote this post on Monday, before Ellsbury was officially the Red Sox's leftfielder. Instead of me hacking it to pieces, just keep that in mind.]
There's a lot of talk about moving Jacoby Ellsbury from center field to left field next season for the Red Sox. With sure-fielding Mike Cameron occupying a roster spot now, the Red Sox have two options: play Cameron in center and move Ellsbury, or keep Ellsbury in his center field spot and slide Cameron over.
The idea behind moving Ellsbury is that his defensive numbers would improve in left field because he has less area to cover because a) he's playing in front of the Monster and b) Cameron is rolling territory to his left.
But there's also the idea that playing in left field in Fenway skews defensive numbers because of the Monster. Since we can't really put a number to the difference between left field in Fenway and a left field in, say, Yankee Stadium, all we can really do is get a rough idea.
So, is there a significant difference?
I decided to look at all the UZR/150 data I could find going back for Red Sox left fielders. Unfortunately, since UZR/150 is a very new statistic, I could only go back until the 2002 season. Meaning we have two full-time left fielders to look at: Jason Bay and Manny Ramirez.We'll start with Bay. Here's how his numbers look in his time in Fenway playing left field:
| INN | UZR/150 | |
|---|---|---|
| 2009 | 1279.1 | -11.2 |
| 2008 | 423.1 | -24 |
Bay improved in his second year with the Sox. This could be for any number of reasons, but one reason may because he became more familiar with Fenway Park. Still, no matter how you slice it, Bay was a bad left fielder for the Sox in 2009.
Here are Manny's numbers with the Red Sox. I only used seasons in which Manny played the most games in left. In 2002, he played a chunk of games as the designated hitter, but he still played more games in left for the Sox than any other player:
| INN | UZR/150 | |
|---|---|---|
| 2008 | 537.2 | 4.8 |
| 2007 | 994.2 | -28.3 |
| 2006 | 1031.1 | -24.2 |
| 2005 | 1225 | -15.7 |
| 2004 | 1087.2 | -9.3 |
| 2003 | 1073 | 11 |
| 2002 | 529.1 | -15.6 |
Manny's UZR numbers fluctuate a tad at the beginning of his career with Boston. In 2002 in limited time, he really struggled. This could be to being unfamilar with the Monster. In 2003, his first full year in left, he put up really good UZR numbers. Maybe he was out to prove himself. But after that he was just on a horrible, horrible slide. His UZR/150 got worse and worse and worse and worse. Then, finally, in limited innings in 2008 he put up positive numbers. But does anyone else think that those numbers would drop if he had put in a full season in Boston?
What can we deduce from Bay's and Manny's numbers? They're both bad fielders. Does that tell us anything about Fenway Park? Not even close.
But let's see how these two guys perform when they're not in Fenway.
Bay playing in left field before he was a Red Sox:
| INN | UZR/150 | |
|---|---|---|
| 2008 | 921.1 | -14.4 |
| 2007 | 1237 | -11.4 |
| 2006 | 1373 | 3 |
| 2005 | 1185 | -4.4 |
| 2004 | 963 | -7.2 |
Other than 2006, which is most certainly an outlier, Bay was just as bad with Pittsburgh as he was with the Red Sox. Last time I checked, though, PNC Park doesn't not have a Yellow Monster in left field.
Manny's numbers since he's been in L.A.:
| INN | UZR/150 | |
|---|---|---|
| 2009 | 812 | -15.4 |
| 2008 | 436.1 | -12.8 |
More bad numbers from Manny. His time in Los Angeles hasn't created numbers as bad as his last two seasons in Boston, but run about par for the course for his entire Red Sox career.
Could his somewhat better numbers as a Dodger be because he got away from the Monster? It's possible, but there are too many factors involved to really say one way or another. If that was true, why would Manny go from a good UZR/150 before being traded from the Sox in 2008 to a horrible one with the Dodgers in the same year? Manny also had a lot to prove these last two seasons, one reason why he could go from a career-low -28.3 with the Sox to a -15.4 in 2009 with the Dodgers.
Perhaps it would be easier to tell how the Monster effects fielders if a) there is more data available and b) the Red Sox's left fielders in that time weren't so bad. Bad fielders tend to be bad everywhere they play. Average fielders could be above average in on park or below average in another.
Maybe we won't ever really know what Fenway does to left fielders' gloves. But we should all sleep easy tonight knowing that no matter who plays left field next season -- Cameron, Ellsbury or even Jeremy Hermida -- they all should be better than Bay or Manny. Because, well, that isn't a feat very hard to accomplish.
25 comments
|
0 recs |
Do you like this story?
Comments
Interesting
good post. I agree with the premise (that it is not so clear ho much UZR is skewed by the Monster). Most people seem unaware of the Manny case because in a very small SS they looked at how his UZR was trending in LA (less than his full time in 08) and at first the numbers looked ok. After that they checked out. More than once I have had to point out that indeed Manny’s D stinks..wherever he plays. I think the Monster probably does have some effect, but it is not so obvious that it changes positive gloves into negative ones. Indeed, all players play 81 games away from home anyway! The Sox likely have a very good understanding of how the Monster skews their home brewed defensive metrics since they chart every ball. Thus, they clearly don’t think Bay was any good, and understand that Cameron>> Ellsbury defensively, and thus took measures to correct this.
I should say
however that the idea of moving Ellsbury to left is just that left is much easier to play, and that reads are much easier (think about the camera-eye view that you have on TV that fools you when a player hits one to dead center and it seems harder to tell how well the ball is hit). +/- and one’s eye really shows that Ellsbury’s problem with reads are in/out. He will not have nearly as much of this in LF. Further, guys that can play CF make great LFs because LFs are bad fielders-things like UZR are weighted to the average at that position (as they should be-one is comparing to how many plays are saved above plays made on other teams on average). These things are independent of the Monster.
Ellsbury will be playing left
The Herald reported this morning that the Red Sox have made it official.
"You know you're having a bad day when the fifth inning rolls around and they drag the warning track." - Mike Flanagan, Baltimore Orioles pitcher, 1992.
Amalie Benjamin of the Globe concurs.
"You know you're having a bad day when the fifth inning rolls around and they drag the warning track." - Mike Flanagan, Baltimore Orioles pitcher, 1992.
Right
I actually wrote this on Monday. Forgot to go edit after the news…
SB Nation's Boston Red Sox community:
OverTheMonster.com
Follow Randy on Twitter: @RCBooth
There really needs to be a way to do home and away splits
What would really make the argument is if there were some way to divide the statistics between playing at Fenway and playing elsewhere for the same player in the same season. I would expect to see a skew for all of the outfield positions. It always seems that there are more problems at Fenway because of strange bounces, the scoreboard, the triangle, etc., that don’t happen as often when the Sox are on the road.
Also—doesn’t the definition of UZR suggest that anything that bounces off a wall is “catchable?” If so, any left fielder will get screwed by the Monster.
That's the "Manny effect"
The UZR inventor admits its a flaw (can’t look it up right now, but I’ve cited it in the past) an says he’s working on it.
I agree home/road splits would be a good thing, but it’d have to be over a number of years to get a clear sense of te Fenway effect.
Manny ain't the only bad man.
Also
Isn’t UZR only good over a three year period? So somebody would have to been with the Sox for 6 years to really get enough innings to determine an accurate “Fenway UZR”, right?
The more data the better
especially for outfielders, since they get fewer chances. Studies show that UZR is not as prone to fluctuations as some think, but indeed more porne to fluctuations than, say, OBP. Part of this is a problem with UZR, and part has to simply do with chances.
But
the numbers so far don’t make a compelling case. Of course the data is poor-only 2 LFs to look at, etc, but Randy’s point is that there seems no clear difference with Manny in LA or Boston, and Bay in 07/08 in Pittsburgh (after knoee injury) and in Boston. This is not to say there is no issue, or that UZR does not have other flaws, but the point should be pondered.
But
The Wall will only hurt you if you have good enough range to make the catch when the Wall isn’t there. Ellsbury certainly does, at least in terms of speed. And I can think of a handful of catches that he’d have made last season, but for the Wall.
Manny, on the other hand, is terrible wherever he goes.
Manny ain't the only bad man.
no evidence whatsoever
and anyway that should be given back by the triangle and RC. Add the fact that Ellsbury is afraid of any outfield wall (remember his misplays in Minn and LA due to Abreu-like wall fear?) makes this a dubious assertion. Plus, Manny supposedly played shallow in Fenway and would have taken balls away that are also not in his “zone” but still his numbers are the same.
Common sense.
If your range is larger than the small confines on Fenway, then Fenway would prevent you from getting to balls that you would get to in a larger park. If Fenway’s dimensions match your range, then you aren’t limited by the field.
But yes, the statisticians haven’t caught up to me yet, so I have no evidence.
Manny ain't the only bad man.
Yet there is no evidence that Fenway hurts UZR
That either means that A) your common sense is wrong, or B) the effect just isn’t significant.
disagree
we just have not enough data to back his common sense up. The 2 LF who played in Fenway are both 2 of the category smaller range so they dont get hurt by the small LF.
by German Red Sox Fan on Jan 9, 2010 6:57 AM EST up reply actions
MGL applies park adjustments
So it’s also possible that he’s adjusting too much. Right now, nobody knows how Fenway affects UZR, and pretending otherwise is disingenuous.
I don't think so.
For some people, the “Monster Effect” is a facile argument to defend the defense of Manny Ramirez and Jason Bay, which “couldn’t possibly be that bad.” According to this line of thinking, Manny and Bay are average fielders who are just being hamstrung by the Monster.
Unfortunately, this is delusional thinking. Bay and Manny are absolutely atrocious fielders. You could take out the Monster, expand the field outward several yards, and that would just be more distance that they can’t cover. Whatever effect the Monster may have, it’s marginal, and certainly not enough to redeem such forsaken defenders.
"It's just a tiny little nick, but it hurts when I get champagne in there."
- Jason Bay, on getting spiked scoring the winning run in ALDS Game Four.
I like
how the Mets scouts just proclaimed Bay is a better fielder than Holliday.
Holliday dropped a fly ball!
In the playoffs! On national television! He sucks! Rawr!
Strikeouts are boring- Besides that, they're fascist. Throw some ground balls - it's more democratic.
by CasanovaWong on Jan 6, 2010 10:56 AM EST up reply actions
No, he caught it...
…just not so much with his hands
Rock me, sexy Jesus...
by nuthinboutnuthin on Jan 6, 2010 11:45 AM EST up reply actions
In the End
“Does the Green Monster Skew Fielding Statistics?”
Yep. It also skews hitting statistics.
In the end everything skews UZR ratings. I skew the ratings.
Slightly Off Topic: Is the formula for UZR posted anywhere online or is it still being treated like the recipe for Coca Cola?
I do think that Fenway has to cause SOME skew for LF UZR
but I don’t think the wall is the main culprit.
I think the lack of foul territory along the deep left line has to hurt UZR. While fly balls hit foul there are not recorded as chances, in other parks those would be mostly easy put outs. So while they don’t ‘hurt’ the Red Sox LF UZR, they aren’t helping, as they would in other parks.
Doesn’t change the fact that Bay and (especially) Manny, are poor fielders. And I don’t think the skew is probably worth more than 1 or 2 per 150. I just seriously doubt it is going to account for the -10 to -20 or worse numbers we saw out of those guys.
Besides – that ’wall skew could easily be balanced by the fact that the Left Fielder is simply responsible for such a smaller area. Some chances that would belong to a LF in another park are either out of play or belong to the CF in Fenway.
OFF TOPIC (tangentially related to valuing defense)
I want to like Bill Simmons (although less than I did), but listening to him talk about baseball with his friends is like watching retarded monkeys trying to prepare a meal. He’s like a single-digit percentage better than Joe Morgan…
Rock me, sexy Jesus...
by nuthinboutnuthin on Jan 6, 2010 3:45 PM EST reply actions

by 




























