Worst Defense? (split from Worst Mets pitching staff ever?)

[Edgy MD]

There's division. That's what makes the start of this thread remarkable. The Mets sucked by all measures. I suppose that's in part due to the greater playing surface. The pitcher gets blamed for a 390-foot homer. The fielder gets statisically hung some for the same 390-foot drive that bounces for a double.

But still!

[Ceetar]

Edgy DC wrote: There's division. That's what makes the start of this thread remarkable. The Mets sucked by all measures. I suppose that's in part due to the greater playing surface. The pitcher gets blamed for a 390-foot homer. The fielder gets statisically hung some for the same 390-foot drive that bounces for a double. But still!

There's plenty of division, but I'm skeptical of _all_ of it. It only takes one small misconception or oversight to totally skew what a stat is saying. Not to say the Mets defense was particularly good. But one of the problems with defensive sabermetrics has always been that they generally represent a pretty small sample size to really judge anything by. Most plays, the routine fly balls, the grounders to second, even I could make. A bad week could ruin the year. A guy could play with a sore calf and play poorly on defense for a week or two because he can't pivot around the bag and he's penalized for it, but does that mean he's a bad fielder? I'm extremely confused about what some of these defensive metrics are trying to tell me.

There was a great article out there recently about the Brewers, who were really bad in 2010 and not so much last year, despite much of the same cast. Defensive positioning.

[Edgy MD]

Ceetar wrote: Edgy DC wrote: There's division. That's what makes the start of this thread remarkable. The Mets sucked by all measures. I suppose that's in part due to the greater playing surface. The pitcher gets blamed for a 390-foot homer. The fielder gets statisically hung some for the same 390-foot drive that bounces for a double. But still! There's plenty of division...

Can you show this to me? Because the data I've seen concurs that the Mets were the worst fielding team in the majors by each of the three leading measures. Concurrence. Corroboration.

[smg58]

The +/- is determined by the percentage of plays that are made over a given distance (the plays are broken into bins by distance and, I believe, direction) league-wide, how many plays an average fielder would have made give the number of grounders or flies hit to a given bin under a particular fielder's watch, and how many plays the fielder actually made. In that setup (UZR is a little different, but I'm not sure how or where), the space of the outfield may magnify a negative or positive reading, but it won't make an above-average fielder look below average. (Playing half your games at an altitude might, however, but that's a discussion which doesn't directly affect the Mets.) At any rate, there should not be a significant park effect for infielders, at least that I'm aware of (without artificial turf affecting things anymore).

Reyes certainly passed the eyeball test, but Jeter has made a career out of doing that, and I saw Elvis Andrus make a number of plays in the postseason that I don't think Reyes makes.

On the whole, there wasn't that much disagreement between UZR and +/- regarding the Mets, with the exception being Angel Pagan. UZR had him going from extremely positive in 2010 to extremely negative in 2011, while the +/- had him going from solidly positive to a -1. The eyeball test certainly had me thinking he was a step or two slower than last year, but the UZR difference seems extreme to me.

[Edgy MD]

My eyeball test had me thinking there was more than a step or two difference. My eyeball test had me thinking he went from excellent to fair.

[metsmarathon]

Ceetar wrote: But one of the problems with defensive sabermetrics has always been that they generally represent a pretty small sample size to really judge anything by. Most plays, the routine fly balls, the grounders to second, even I could make. A bad week could ruin the year. A guy could play with a sore calf and play poorly on defense for a week or two because he can't pivot around the bag and he's penalized for it, but does that mean he's a bad fielder?

isn't that true of any stat that accumulates over the course of an entire season? a bad outing can blow up an ERA something fierce, or drain a batting average.

plus, and i think this is key for most statistics, they measure performance, not talent. and past performance is not always indicative of future results.

[Ceetar]

There's division. That's what makes the start of this thread remarkable. The Mets sucked by all measures. I suppose that's in part due to the greater playing surface. The pitcher gets blamed for a 390-foot homer. The fielder gets statisically hung some for the same 390-foot drive that bounces for a double. But still! There's plenty of division... Can you show this to me? Because the data I've seen concurs that the Mets were the worst fielding team in the majors by each of the three leading measures. Concurrence. Corroboration.

was agreeing with you in re: fielding metrics. I then was trying to say that three wrongs don't make a right.

[Benjamin Grimm]

But three lefts do make a right.

[Ceetar]

metsmarathon wrote: Ceetar wrote: But one of the problems with defensive sabermetrics has always been that they generally represent a pretty small sample size to really judge anything by. Most plays, the routine fly balls, the grounders to second, even I could make. A bad week could ruin the year. A guy could play with a sore calf and play poorly on defense for a week or two because he can't pivot around the bag and he's penalized for it, but does that mean he's a bad fielder? isn't that true of any stat that accumulates over the course of an entire season? a bad outing can blow up an ERA something fierce, or drain a batting average. plus, and i think this is key for most statistics, they measure performance, not talent. and past performance is not always indicative of future results.

Certainly, but if the stat isn't predicting future results at all, is it really measuring what we think it's measuring? UZR is one of those, that drastically swings year to year sometimes for any number of reasons both known and unknown. We all watched Pagan look worse defensively, but sometimes the defender looks roughly the same and there is still a swing. Pagan will be a very interesting case next year.


It is true of any stat, but fielding is the least difficult to actually do and probably affected by the most random of effects. wind speed, ambient light, crowd noise (and color! one patch of white-shirted fans in the upper deck at just the right angle and..) , how wet the grass is, positioning, quality of the other fielders (did Pagan go less aggressively towards right field due to some subconscious confidence that Beltran would get everything cause Beltran is awesome and his idol?) How do you measure any of that? How do you know you're measuring actual talent versus merely luck-driven results?

[metsmarathon]

i think that every stat in baseball, and sports in general, measures, and is dominated by, luck-based results.

[Ceetar]

metsmarathon wrote: i think that every stat in baseball, and sports in general, measures, and is dominated by, luck-based results.

But we've gotten a lot better, particularly with hitting, at figuring out what's luck and what's not.

Also, we don't even know what % of the game fielding really is. Reyes for instance had already created more runs above average batting in April than he and Turner supposedly cost in defense all year.

[Edgy MD]

There's division. That's what makes the start of this thread remarkable. The Mets sucked by all measures. I suppose that's in part due to the greater playing surface. The pitcher gets blamed for a 390-foot homer. The fielder gets statisically hung some for the same 390-foot drive that bounces for a double. But still! There's plenty of division... Can you show this to me? Because the data I've seen concurs that the Mets were the worst fielding team in the majors by each of the three leading measures. Concurrence. Corroboration. was agreeing with you in re: fielding metrics. I then was trying to say that three wrongs don't make a right.

I disagree. Agreement from three independent sources certainly make something righter. That's the point of noting corroboration. And they're all working from a very large data set --- fielding by an entire team at all positions.

[TransMonk]

Jose getting written up for his defense:

http://espn.go.com/blog/statsinfo/post/_/id/30796/jose-reyes-aramis-ramirez-share-flaws

[Guest attgig]

TransMonk wrote: Jose getting written up for his defense: http://espn.go.com/blog/statsinfo/post/_/id/30796/jose-reyes-aramis-ramirez-share-flaws

Last year, what hurt Reyes was that he struggled on balls hit to the shortstop-third base hole. He made 18 fewer plays than the average shortstop on balls hit in that area. On all other balls hit to him, Reyes rated well above average in getting outs.

could that be because Wright has such horrible range, and the balls hit to that general area - most good fielding 3bmen would be getting and not leaving it to the SS?

[Edgy MD]

If true, that should reflect statistically on Wright, not Reyes.

[Ceetar]

attgig wrote: Jose getting written up for his defense: http://espn.go.com/blog/statsinfo/post/_/id/30796/jose-reyes-aramis-ramirez-share-flaws Last year, what hurt Reyes was that he struggled on balls hit to the shortstop-third base hole. He made 18 fewer plays than the average shortstop on balls hit in that area. On all other balls hit to him, Reyes rated well above average in getting outs. could that be because Wright has such horrible range, and the balls hit to that general area - most good fielding 3bmen would be getting and not leaving it to the SS?

No, it could reflect positively on Wright/Turner/Murphy, although I don't think it does. If the 3B makes MORE plays, than the SS would end up making less. I don't know the details of those stats. Are those 18 fewer plays hits, or simply balls in play? What's the percentage? It can't simply mean 18 because Reyes did play fewer games (I think anyway) than the average SS. Is it prorated for 162?

I always wonder how much of this is positioning. Was Reyes positioned more towards 2B to account for what they may have perceived as lesser range there?

[Edgy MD]

Well, if his positioning was wise, I imagine it would have shown up statistically.

[Ceetar]

Edgy DC wrote: Well, if his positioning was wise, I imagine it would have shown up statistically.

Oh, I was pondering the opposite actually. That his positioning was unwise. More specifically, are the Mets correctly aware of the ranges of their infielders to the point of positioning them properly? You'd hope so, but I'm not sure it's that simple.

[Edgy MD]

Well, they've fired the infield guy on the coaching staff. But Reyes has been in the division as long as anybody not named Chipper. He should know where to sit and when.

It's a theory. I'm impressed on how pro-active the Mets were with the coaching staff cuts.

[Ceetar]

Edgy DC wrote: Well, they've fired the infield guy on the coaching staff. But Reyes has been in the division as long as anybody not named chipper. He should know where to sit and when.

Hmm..maybe. I imagine players tend to give more credit to their peers than is perhaps warranted. I get he may know Wright, but Murphy (on either side?) and Turner? Tejada? Would he 'shake off' a positioned called from the bench anyway?

[Edgy MD]

I'm certain it's a fool's game to try and stick Reyes' failures, such as they were, on the guys on the left and right. The statistics are created to operate independent of that.