Ellsbury's Inexplicable RngR

by SBGonzalez on Oct 31, 2009 11:01 PM EDT Comment 26 comments

For those uninformed: Jacoby Ellsbury is a godawful center fielder, according to UZR, Fangraphs' fielding statistic of choice. UZR has Ells at 16.5 runs below average in 153 games at center in '09, making him one of 3 CF with a UZR below -10.0, the others being Dexter Fowler and Vernon Wells. Ellsbury's noodle arm didn't exactly help him out here, but the main reason Jacoby's UZR was so poor was his -14.0 RngR, or Range Runs Above Average. The first time I saw that number, I was pretty much shocked; how could such a fast guy have such poor range?

UZR tends to have really weird fluctuations; observe Raul Ibanez's UZR in LF go from okay to good to bad to absolutely horrible to good again. It's also worth noting that Ellsbury posted a positive RngR in 66 games at center last year. So, is Ellsbury really this bad at center, or can we chalk this up to wonky UZR data/Jason Bay having zero range/Julio Lugo?

0 recs | Comment 26 comments

Story-email Email Printer Print

Comments

Watch when the ball first comes off the bat.

He routinely makes bad reads, which he has to make up for with speed. Most of his diving catches shouldn’t have needed a diving attempt in order for the fielder to make the play.

@bs_uf15bosox9be 12 pieces of bacon, a Red Bull, and go get 'em; Learn to use SB Nation

by bs.uf15bosox9bears23 on Nov 1, 2009 12:20 AM EDT reply actions 0 recs

True, I've seen him make more than a few bad reads.

But how does that translate into his UZR data from last year, where he was good in CF and absolutely insane in LF (where I’d think reaction time would count more, due to the small size of Fenway’s LF)?

by SBGonzalez on Nov 1, 2009 1:00 AM EDT up reply actions 0 recs

Because he didn't get to play nearly as much last year?

So the scale’s out of whack. To be honest, UZR is best used with at least two to three seasons worth of data.

@bs_uf15bosox9be 12 pieces of bacon, a Red Bull, and go get 'em; Learn to use SB Nation

by bs.uf15bosox9bears23 on Nov 1, 2009 1:03 AM EDT up reply actions 0 recs

It confuses the hell out of me too.

I know people say he gets bad reads, but would anyone be surprised if he posted a positive UZR next year? I would not at all. Coco fluctuated 20 runs from one year to the next . These huge fluctuations really make me question UZR for OFer’s.

"Ninety percent [of my salary] I'll spend on good times, women, and Irish whiskey. The other ten percent I'll probably waste."
-Tug McGraw

by BTLove on Nov 1, 2009 1:32 PM EST reply actions 0 recs

I'm going to predict that he's still below average, but not by double digits runs.

@bs_uf15bosox9be 12 pieces of bacon, a Red Bull, and go get 'em; Learn to use SB Nation

by bs.uf15bosox9bears23 on Nov 1, 2009 8:13 PM EST up reply actions 0 recs

UZR

I dont think UZR means anything for outfielders. Everyones fluctuates so much that i dont think it really has much weight. Even though a lot of people say he gets bad reads, i have seen many balls hit that i dont think Ellsbury has a chance of getting to and he makes a diving catch. So saying all of his dives would be caught by other CFs is false

by cnubsbl16 on Nov 6, 2009 9:07 AM EST up reply actions 0 recs

Of course, ALL of his dives wouldn't be caught.

There are plenty of time when it appears Ells makes impossible plays, but that may have been caused by the poor route he started on before the camera cut.

@bs_uf15bosox9be 12 pieces of bacon, a Red Bull, and go get 'em; Learn to use SB Nation

by bs.uf15bosox9bears23 on Nov 6, 2009 4:06 PM EST up reply actions 0 recs

You make it seem like all of his dives are not important because he took a bad route. I guarantee you he can make plays that many other CFs cannot make

by cnubsbl16 on Nov 8, 2009 10:14 AM EST up reply actions 0 recs

not according to UZR.

"Ninety percent [of my salary] I'll spend on good times, women, and Irish whiskey. The other ten percent I'll probably waste."
-Tug McGraw

by BTLove on Nov 8, 2009 6:11 PM EST up reply actions 0 recs

i dont believe in UZR….especially for an outfielder

by cnubsbl16 on Nov 9, 2009 7:57 AM EST up reply actions 0 recs

... Why not?

We can’t make the argument for you. Now excuse me, I need to go strangle myself for quoting my least favorite teacher.

@bs_uf15bosox9be 12 pieces of bacon, a Red Bull, and go get 'em; Learn to use SB Nation

by bs.uf15bosox9bears23 on Nov 13, 2009 7:47 PM EST up reply actions 0 recs

Unless you can get Beltran for a modest price (unlikely), I think it’s best to pursue a bat for LF (Bay or Holliday) and ride Ellsbury in CF. I’m hoping that he’ll improve his reads and become the defensive CF that everyone hopes he will be.

by mg050369 on Nov 1, 2009 3:16 PM EST reply actions 0 recs

Does anybody have a home/away UZR rating?

I’d be interested to see how much of an effect Fenway has on the CF’s UZR ratings. . . .

by lone1c on Nov 5, 2009 12:50 AM EST reply actions 0 recs

can't find that anywhere

but I did notice that every year his UZR/150 has changed by at least 17. That means we’re in for him having a great defensive year in ’10!!!!!

by wolf9309 on Nov 5, 2009 9:54 AM EST up reply actions 0 recs

excuse the sarcasm

I’m just of the belief that as amazing as today’s intensive stats are, UZR is very misleading. I think the individual sections that comprise UZR are worth looking at, but in combination with watching the games because the system does not seem very accurate. The fluctuations year to year with some players are just ridiculous.

Maybe Ellsbury has contacts and forgot to wear them all of this year and ’07?

by wolf9309 on Nov 5, 2009 9:57 AM EST up reply actions 0 recs

Just go look at Melky Cabrera's stats

They’ll show you that UZR, etc. really have more to do with circumstance than talent. Melky has a weak arm but is generally agreed upon as being a decent center fielder, in spite of a lower FP this year than Ellsbury.

Key differences? Jeter stepped up his defense at short, Cano isn’t lazy anymore, and Damon covers more ground faster than Bay does.

Coco played a great CF in spite of the defensive black hole in #24. Amazing stats in ’07 yet not so great ones in ’06 or ’08. Is he really just a mediocre defender who happened to have everything go right for an entire year? Or should we take these states with an entire bucket of salt?

by AustinRedSox on Nov 5, 2009 12:07 PM EST reply actions 0 recs

I've always thought the same thing.

But recently I read on FG’s that UZR fluctuates no more than the most reliable offensive stats. Still, I think the Fenway OF has some strange things going on.

"Ninety percent [of my salary] I'll spend on good times, women, and Irish whiskey. The other ten percent I'll probably waste."
-Tug McGraw

by BTLove on Nov 5, 2009 12:44 PM EST up reply actions 0 recs

That's probably the cause of some of the fluctuation.

@bs_uf15bosox9be 12 pieces of bacon, a Red Bull, and go get 'em; Learn to use SB Nation

by bs.uf15bosox9bears23 on Nov 5, 2009 9:11 PM EST up reply actions 0 recs

The thing with that is...

UZR, which is primarily based on the range of a player, should not fluctuate as much as offensive stats, even the more reliable ones like wOBA, which circumstance plays a part in. Range should remain constant, at least for a young guy like Ells.

by SBGonzalez on Nov 5, 2009 10:26 PM EST up reply actions 0 recs

Why?

Why should stats that indicate a player’s abilty to cover ground be more consistent than stats that indicate a player’s ability to hit a baseball?

"Ninety percent [of my salary] I'll spend on good times, women, and Irish whiskey. The other ten percent I'll probably waste."
-Tug McGraw

by BTLove on Nov 6, 2009 3:23 AM EST up reply actions 0 recs

Aside from the obvious issues with aging or changing techniques...

UZR is also a counting stat, so less opportunities will lessen the chance that 1) the extrapolated UZR is accurate, and 2) the balls were distributed normally.

@bs_uf15bosox9be 12 pieces of bacon, a Red Bull, and go get 'em; Learn to use SB Nation

by bs.uf15bosox9bears23 on Nov 6, 2009 4:09 PM EST up reply actions 0 recs

If I recall

The idea with UZR is it is supposed to be over huge sample sizes, I believe 3 year windows. Since Ells has only played one year as full time, not accurate yet!

by wolf9309 on Nov 7, 2009 11:48 AM EST up reply actions 0 recs

which makes it kind of useless

except for making decisions about players in retrospect or after they’ve been in the majors for a long time. Maybe I’m wrong about the 3-year window, but I seem to remember that.

by wolf9309 on Nov 7, 2009 11:49 AM EST up reply actions 0 recs

Yeah, 2-3 years would be great.

But if you can combine it with scouting, it can be useful. It’s creator, MGL, says that we should think of it as a score of how well they performed by a standard set by their peers. Not how gifted defensively they are, not how much better they are than before (I’ve seen that interpretation somewhere), etc. It’s still a work in progress, but it’s far and away the best freely available stat.

@bs_uf15bosox9be 12 pieces of bacon, a Red Bull, and go get 'em; Learn to use SB Nation

by bs.uf15bosox9bears23 on Nov 13, 2009 7:51 PM EST up reply actions 0 recs

Pardon the innacurate language, I know the hitting stats and methods well, but am not so up on the defensive ones:

I read someplace about 10 days ago that Ells’s problems come with ball that are hit shallow. If that is true, then it is additional singles he is giving up – so the BA/OBP Runner moving up/scoring gets worse. I can’t rember any bloops getting by him, just landing in front.

Now I can also remember huge range to the left and right, and back, robbing doubles, triples – more than an average CF would. So that would depress a bit (but not enough the BA/OBP), but what about the SLG? is that factored into UZR? I thought it was like 84 grids that were examined.

If Ellsbury let up 11 extra singles last year, that’s not good, but if he caught 7 balls for 15 TB’s that an average CF wouldn’t get, would that simply be seen as a net 4 catches not made?

by Tony Horton on Nov 5, 2009 1:12 PM EST reply actions 0 recs

very good thought

but yeah, that is factored in. There’s 64 zones UZR uses I believe- the plays are judged based on the average run value of a hit into that particular zone.

One thing I don’t completely understand and need to read up on are the park factors. My impression is that at a park like fenway with the short, high left field wall, it’s not factored very accurately, but like I said, I need to read up on that more.

If you go to http://www.fangraphs.com/blogs/index.php/glossary/ and look at the 2 part primer on UZR, it’s pretty informative.

by wolf9309 on Nov 5, 2009 2:58 PM EST up reply actions 0 recs