Does this make sense based on what you've seen this year?

[Can_of_corn]

They don't feel it's worth discussing because it calls into question the whole sabermetrics calculations. They can't quantify how Nick could have a -8.9 after only 12 games in which he has actually played well defensively. So saying it's a SSS and not worth discussing pushes off the discussion until the number of games and the plays in those games is too numerous to remember them all. So it's easier to verify he's played poorly and deserves his ranking.

No, we don't feel like it is worth discussion because it is analogous to making a thread about Davis not hitting for power after the first week of the season.

[blid]

No, we don't feel like it is worth discussion because it is analogous to making a thread about Davis not hitting for power after the first week of the season.

It's not analogous to that at all... Davis didn't hit for heavy power the first week of the season. That's a fact. I feel like Frobby explained this. Drawing conclusions based on a SSS is a mistake, like saying that Davis won't hit for power this season or that Nick will be poor defensively this season. But saying that Davis had one HR so far is a verifiable fact. Markakis has a busted UZR so far this season is a fact. The point of the thread is to figure out why he would have that busted UZR, not to make judgments about Markakis's defensive abilities and outlook for the season. Have there been balls he hasn't been getting to? Where does this negative UZR come from? That's the root question trying to be dug out.

It's idiotic. He has 11 games in RF and 10 starts. He has 24 total chances. These stats need 3-4 years to approximate the stability of one years worth of offensive stats. He has less that 1/2 of 1% of that sample. DRS says he is 1 play below average on 24 chances which gives him a rate of minus 13 runs below average. The rate stat is useless at this point.

OK, he's had 24 total chances. Agreed that the UZR number itself tells us nothing about Nick, with it being determined based on such a small sample. But can we figure out which of those chances he fielded in a subpar manner? Or which chances UZR thinks, anyway, he fielded poorly in order to arrive at that number?

[Ohfan67]

No, we don't feel like it is worth discussion because it is analogous to making a thread about Davis not hitting for power after the first week of the season.

It's absolutely not the same. Davis did not hit any home runs. Thats a fact. Dumb to project based on SSS, but still a fact. Frobby is basically saying that Markakis has looks good in the field but the defensive metric says he sucks. Yes, small sample size and worthless for projection but Frobby is basically asking why would the system seem so misaligned with Markakis' actual play to date.

[CA-ORIOLE]

It's not analogous to that at all... Davis didn't hit for heavy power the first week of the season. That's a fact. I feel like Frobby explained this. Drawing conclusions based on a SSS is a mistake, like saying that Davis won't hit for power this season or that Nick will be poor defensively this season. But saying that Davis had one HR so far is a verifiable fact. Markakis has a busted UZR so far this season is a fact. The point of the thread is to figure out why he would have that busted UZR, not to make judgments about Markakis's defensive abilities and outlook for the season. Have there been balls he hasn't been getting to? Where does this negative UZR come from? That's the root question trying to be dug out.

OK, he's had 24 total chances. Agreed that the UZR number itself tells us nothing about Nick being determined based on such a small sample. But can we figure out which of those chances he fielded in a subpar manner? Or which chances UZR thinks, anyway, he fielded poorly in order to arrive at that number?

i think DRS or UZR publish charts at the end of the year. There were pretty detailed ones comparing Jones and Trout in previous years as I recall. The Fielding Bible lays it out in tabular format but you have to be a subscriber or get the book at the end of the year to get the rating data. A small percentage of plays make the difference. Figure 80% (maybe more of these chances are routine), So at this point we're probably talking about 4-5 "measurable" plays that are separating outfielders right now. Not only that, this is a dynamic structure so a play that was made earlier in the year may be worth more or less at the end of the year as the sample size increases.

[Can_of_corn]

It's not analogous to that at all... Davis didn't hit for heavy power the first week of the season. That's a fact. I feel like Frobby explained this. Drawing conclusions based on a SSS is a mistake, like saying that Davis won't hit for power this season or that Nick will be poor defensively this season. But saying that Davis had one HR so far is a verifiable fact. Markakis has a busted UZR so far this season is a fact. The point of the thread is to figure out why he would have that busted UZR, not to make judgments about Markakis's defensive abilities and outlook for the season. Have there been balls he hasn't been getting to? Where does this negative UZR come from? That's the root question trying to be dug out.

OK, he's had 24 total chances. Agreed that the UZR number itself tells us nothing about Nick, with it being determined based on such a small sample. But can we figure out which of those chances he fielded in a subpar manner? Or which chances UZR thinks, anyway, he fielded poorly in order to arrive at that number?

It's absolutely not the same. Davis did not hit any home runs. Thats a fact. Dumb to project based on SSS, but still a fact. Frobby is basically saying that Markakis has looks good in the field but the defensive metric says he sucks. Yes, small sample size and worthless for projection but Frobby is basically asking why would the system seem so misaligned with Markakis' actual play to date.

It is the same.

Defensive metrics take TWO SEASONS to reach a point where the data sample is large enough. 12/2=6. Talking about 12 games worth of defensive metrics is the same as talking about six games of offensive data.

[CA-ORIOLE]

It is the same.

Defensive metrics take TWO SEASONS to reach a point where the data sample is large enough. 12/2=6. Talking about 12 games worth of defensive metrics is the same as talking about six games of offensive data.

Most of the stuff I've read says average 3-4, but yeah.

[mark_beckens]

No, we don't feel like it is worth discussion because it is analogous to making a thread about Davis not hitting for power after the first week of the season.

You can't explain it. So you hide behind SSS. What has Nick done to warrant a -8.9? He must have played poorly to garner such a poor rating. You should be able to explain how.

[Can_of_corn]

You can't explain it. So you hide behind SSS. What has Nick done to warrant a -8.9? He must have played poorly to garner such a poor rating. You should be able to explain how.

The one ball Frobby is referring to, over the course of a 12 game sample size, could be enough to skew his numbers. Almost all outfielders make the plays they are supposed to, you miss just a few of them, and the rating is going to suffer.

[Lester Freamon]

Nick Markakis has a -8.9 UZR/150 through 12 games. Thinking back on the season so far, I can only think of one ball he could have caught, but didn't -- a diving play where he gloved it but the ball escaped his mitt when he hit the ground. Other than that, he's looked great playing the RF wall, has thrown out a runner at 2B, and made one really nice diving play (the game before the one he missed). It's things like this that make me wonder how these stats are designed. That said, I haven't watched every minute of every game, so if there are other plays Nick could have made but didn't, I'd like to hear it.

I'll try and take a stab at it in, what I hope, is the spirit in which you asked. I think it's hard for us to assess someone defensively because, and in particular someone like Markakis, because only see the plays he makes and we may tend to assume that the plays he didn't make wouldn't have been made by someone else. I think unless you are looking at video of two different players attempting to make the same type of play, whether on overlay or side by side, it's can be hard to judge. Add to that the Nick, at least IMO other may disagree, never really looks bad, he always seems to be in proper position to get to balls that he doesn't catch and seems to play caroms very well.

Nelson Cruz, OTOH (and I realize that his UZR numbers career wise, don't know about thus far this year, are similar if not better than Nick's), often looks bad or uncomfortable in the OF even on plays that he makes. It is quite easy for me to imagine David Lough in Cruz's place and making the play on at least 3 balls hit to Cruz in LF this year so far that I have seen. And although I imagine Lough would play a much better RF than Nick, it is not as easy to visualize as it is with Cruz in LF.

Does that make any sense? Long story short, even though I agree that it looks as though Nick has played a solid RF so far this season I can imagine how my eyes might be deceiving me.

[mark_beckens]

The one ball Frobby is referring to, over the course of a 12 game sample size, could be enough to skew his numbers. Almost all outfielders make the plays they are supposed to, you miss just a few of them, and the rating is going to suffer.

So 1 out of 24 = -8.9. Gotcha!

[CA-ORIOLE]

So 1 out of 24 = -8.9. Gotcha!

Minus 1 (approximately) is his counting stat to date. -8.9 is a rate projected over 150 games.

[mark_beckens]

Minus 1 (approximately) is his counting stat to date. -8.9 is a rate projected over 150 games.

If he was to retire right now. His season UZR would be? -8.9 because he failed to make one play.

[CA-ORIOLE]

His UZR would be approximately minus 1. Actually a fraction of minus 1 at this point by UZR. His projected rate would be -8.9 per 150 games. -8.9 is a rate stat. Using small sample size UZR/DRS is dumb. Using small sample rate stats is moronic.

[Stotle]

The hostility in this thread is hilarious. I think the answer has already been stated, but essentially with a small collection of plays one or two balls can vastly skew the rate stat. If Nick were to play the exact same thirteen and a half times it would ultimately come out to around -8.9. That doesn't seem that crazy to me. He'll go through 12 game spurts where his "rate" over that period will be +22, and others where it will be 0 or -15.

The Davis HR was not a great example because it's a counting stat. It's closer to looking at Davis' slugging percentage over a series where he collected a couple of singles and asking if that was a fair representation of his power. Maybe; maybe not. Was he making lots of loud contact and finding gloves, or was he swinging through pitches and generally looking toothless?

Estimating the UZR rating as appropriate or not is a tough exercise to consider over twelve games. Even more so if you enter the exercise with preconceived notions.

[notyourbuddy]

If a stat isn't judging a player properly it shouldn't matter if its over 1 game or 20 the merits of that stat become suspect. Seems worth discussing to me.

(that being said a fly ball to right field just landed 5 inches out of Markakis' reach. I bet a better RF would have caught that)