Defense?

[Frobby]

Errors really don't factor in for UZR and FB +-. It's the % of lpays made by all fielders at that position on balls hit to that particular bucket. It doesn't matter what the official scorer says, either the play was made or it wasn't. As to Ellsbury I think Theo was blowing smoke. 2009 was Ellsbury's first fiull year in CF at Fenway. FB UZR, and, Rtot all agreed he was -8-9 RS that year. I expect they were right. But that doesn't mean he is a poor CF. If his numbers remained- 6-9 RS for the following 2 years it probably would. It most likely means it's tough to learn how to play CF at Fenway. I wonder what players on the O's, the only team you are in any position to judge, besides Jones and Markakis, are not reflected accurately in the defensive metrics. They say Hardy and Wieters are among the very best at their positions Reimold has improved in a SSS, Andino is a little above averge at 2B, Reynold i, horrible at 3B, but ok at 3B, and Tatum is average as a backup catcher. Which of thes assesments do you take issue with? If not, how is it that they, are accurate with 7 players and so far off base with 2?

I'm going to answer you in a hypothetical way. I think different factors are in play at different positions on the field. Probably the greatest variation is for outfielders due to the significant differences in where the fences are, how high they are, etc. Infielders all play on infields of the same dimension, though the speed of the infield and the extent of foul territory can vary. For catchers, probably the single biggest variable is how good the pitchers are at holding runners. I don't know enough about how the different systems account for these differences to say whether there are problems at certain positions that don't exist at others. And the creators of these systems don't do a good job of explaining it.

The Markakis home/away splits really bother me, because they are contrary to common sense and contrary to what I see. Common sense tells you that an outfielder will usually play better at home than he does on the road, because he is so familiar with the dimensions of his home ballpark. And that's what I see, watching the games. So, when I see that Markakis has better numbers on the road every year than he does at home, that tells me something is wrong, and pro-Oriole bias has nothing to do with that.

[El Gordo]

Oh's D by position according to UZR:

1B -3.7(Lee 1.4, Reynolds -5.3, Davis -2.5)

2B -3.3

3B -26.7

SS +12.2

LF -15.3(Reimold 0.9, Pie -13.4)

CF -12.9

RF -4.0

C (FB#'s)+14 RS.

From this 3B, LF, and CF, are top priorities. LF is a question depending on how much Reimold's improvement in a SSS is real.

[Dipper9]

Oh's D by position according to UZR:

1B -3.7(Lee 1.4, Reynolds -5.3, Davis -2.5)

2B -3.3

3B -26.7

SS +12.2

LF -15.3(Reimold 0.9, Pie -13.4)

CF -12.9

RF -4.0

C (FB#'s)+14 RS.

From this 3B, LF, and CF, are top priorities. LF is a question depending on how much Reimold's improvement in a SSS is real.

I am not going to pretend to have ANY idea how they come up with these numbers, but this is where my eyes disagree with those numbers. I think they put too much emphasis on errors. As an example, I saw Fielder unable to scoop a low throw last night that Reynolds was scooping with his eyes closed. But yet those errors he made at first lowered his rating. I know it was just one play, but watching Reynoldsat first, I think he was pretty darn good, and I'd go so far as to say he saved a whole lot more errors with his scooping than errors that he actually had.

[Frobby]

I am not going to pretend to have ANY idea how they come up with these numbers, but this is where my eyes disagree with those numbers. I think they put too much emphasis on errors. As an example, I saw Fielder unable to scoop a low throw last night that Reynolds was scooping with his eyes closed. But yet those errors he made at first lowered his rating. I know it was just one play, but watching Reynoldsat first, I think he was pretty darn good, and I'd go so far as to say he saved a whole lot more errors with his scooping than errors that he actually had.

Here is an interesting article from late August regarding the leading 1B in terms of "scoop percentage." According to this article, the average 1B scoops 82% of throws successfully. Fielder was at 70%. http://www.billjamesonline.com/first_baseman_scoops/

Fangraphs tracks successful scoops by 1B, but not scoop percentage: http://www.fangraphs.com/leaders.aspx?pos=1b&stats=fld&lg=all&qual=y&type=0&season=2011&month=0&season1=2011&ind=0&team=0&players=0

Reynolds had 17 successful scoops in 44 games played at 1B. That's a very high number when you consider that nobody had more than 58 successful scoops over a full season, and the average is in the 30ish range. I don't know how many he missed. However, an average 1B saves 82% so if you assume that Reynolds was at 100%, that means he scooped 3 more than average. At the same time, he made 5 errors at 1B, which is very high (an average 1B would be expected to make 2-3 errors in the same number of chances as Reynolds). So, extra scoops vs. errors is probably a wash at best. But the biggest reason Reynolds' UZR at 1B was low was that his Range Rate was low (-3.8). That makes sense to me, because there were quite a few times when a ball was hit towards the 1B-2B hole where Reynolds retreated to 1B instead of going after the ball in the hole. Considering Reynolds is new to 1B, I'd have some hope that he would improve in that area given more experience.

[wildcard]

The Orioles' third base situation doesn't currently have a solution. The fallback position is to use Reynolds at first and Davis at third, and that's probably worse than the Lee/Reynolds situation they had for most of this year.

Andino may be getting better at second. Left field is still probably an issue. So I don't see any definite positives or gains out of 1B, 3B, LF, CF, and 2B is either a small gain or an unknown.

It seems premature, at best, to assume that defense won't be a major problem in 2012.

Getting Reynolds off 3B is a huge gain IMO. I also think that Andino is likely to be a much improved player defensively when compared on a year by year basis. He has matured and gained experience.

[El Gordo]

I'm going to answer you in a hypothetical way. I think different factors are in play at different positions on the field. Probably the greatest variation is for outfielders due to the significant differences in where the fences are, how high they are, etc. Infielders all play on infields of the same dimension, though the speed of the infield and the extent of foul territory can vary. For catchers, probably the single biggest variable is how good the pitchers are at holding runners. I don't know enough about how the different systems account for these differences to say whether there are problems at certain positions that don't exist at others. And the creators of these systems don't do a good job of explaining it.

The Markakis home/away splits really bother me, because they are contrary to common sense and contrary to what I see. Common sense tells you that an outfielder will usually play better at home than he does on the road, because he is so familiar with the dimensions of his home ballpark. And that's what I see, watching the games. So, when I see that Markakis has better numbers on the road every year than he does at home, that tells me something is wrong, and pro-Oriole bias has nothing to do with that.

I believe this is the crux of the matter were you are concerned and perhaps you aren't the best person for objectivety here. IMO Markakis has a poor first step, and because of this he is slow at getting to balls hit in front of him, in the gap and down the line. In OPACY, because it is shallow, he can't play deep enough to get angles that will allow him to compensate for this. In larger parks he can. CoNsequently he gets to more balls in the gap and down the line, that go for doubles in OPACY. I agree that the defensive metrics do a better job with some positione than others. They are particularly weak with C. Also I don't think they measure 1B as well as they could, because I don't think they give credits for picks. It seems they evaluate a 1B the same as a 3B. As to the OF, they do make adjustments for park differences. It's not a perfect sytem, but is continually being refined, and is better than the alternative, no objective measure at all, just a cacophony of subjectivety.

[El Gordo]

Here is an interesting article from late August regarding the leading 1B in terms of "scoop percentage." According to this article, the average 1B scoops 82% of throws successfully. Fielder was at 70%. http://www.billjamesonline.com/first_baseman_scoops/

Fangraphs tracks successful scoops by 1B, but not scoop percentage: http://www.fangraphs.com/leaders.aspx?pos=1b&stats=fld&lg=all&qual=y&type=0&season=2011&month=0&season1=2011&ind=0&team=0&players=0

Reynolds had 17 successful scoops in 44 games played at 1B. That's a very high number when you consider that nobody had more than 58 successful scoops over a full season, and the average is in the 30ish range. I don't know how many he missed. However, an average 1B saves 82% so if you assume that Reynolds was at 100%, that means he scooped 3 more than average. At the same time, he made 5 errors at 1B, which is very high (an average 1B would be expected to make 2-3 errors in the same number of chances as Reynolds). So, extra scoops vs. errors is probably a wash at best. But the biggest reason Reynolds' UZR at 1B was low was that his Range Rate was low (-3.8). That makes sense to me, because there were quite a few times when a ball was hit towards the 1B-2B hole where Reynolds retreated to 1B instead of going after the ball in the hole. Considering Reynolds is new to 1B, I'd have some hope that he would improve in that area given more experience.

I agree with this on Reynolds. Range is his major problem at 1B, and some of this is a matter of adjusting to a new position. But errors don't really factor in to the equation except as a play not made with a certain % of likelyhood to be made.

[El Gordo]

I am not going to pretend to have ANY idea how they come up with these numbers, but this is where my eyes disagree with those numbers. I think they put too much emphasis on errors. As an example, I saw Fielder unable to scoop a low throw last night that Reynolds was scooping with his eyes closed. But yet those errors he made at first lowered his rating. I know it was just one play, but watching Reynoldsat first, I think he was pretty darn good, and I'd go so far as to say he saved a whole lot more errors with his scooping than errors that he actually had.

So yoy think Reynolds isbeing grossly mischaracte ized here? Which others are blatantly wrong, IYO. Remember these are numbers fo positiions and not individual players. Andino has played some SS and 3B and Reimold has played some RF. Pie played some CF as well as LF, BRob and Adams have played some 2B, and Davis has been playing 3B for the last month.

[CA-ORIOLE]

I believe this is the crux of the matter were you are concerned and perhaps you aren't the best person for objectivety here. IMO Markakis has a poor first step, and because of this he is slow at getting to balls hit in front of him, in the gap and down the line. In OPACY, because it is shallow, he can't play deep enough to get angles that will allow him to compensate for this. In larger parks he can. As to the OF, they do make adjustments for park differences. It's not a perfect sytem, but is continually being refined, and is better than the alternative, no objective measure at all, just a cacophony of subjectivety.

Your explanation is good, but the reality is some of the park differences are probably too difficult to overcome. Maybe RF in OPACY is one of them. The more adjustments you make, the more errors you are open too. Look at Crawford in LF at Fenway. He was a minus 2 after being consistently in the plus 15 range. Maybe it just isn't possible to evaluate LF defense properly in Fenway Park and we may have to leave it at that. Probably there are other such cases. It would help if they showed performance by each player in the respective parks, so to some extent I agree with Frobby about transparency etc.

That being said I don't see this issue as defense only. How does Adrian Beltre increase his OPS+ by 30-40% in hitters parks over SAFECO. Adrian Gonzlaez etc. there are drastic home away splits on offense and big splits in OPS+. The plus after OPS should adjust for park, but in a lot of cases it just doesn't give you the right information.

[El Gordo]

Your explanation is good, but the reality is some of the park differences are probably too difficult to overcome. Maybe RF in OPACY is one of them. The more adjustments you make, the more errors you are open too. Look at Crawford in LF at Fenway. He was a minus 2 after being consistently in the plus 15 range. Maybe it just isn't possible to evaluate LF defense properly in Fenway Park and we may have to leave it at that. Probably there are other such cases. It would help if they showed performance by each player in the respective parks, so to some extent I agree with Frobby about transparency etc.

That being said I don't see this issue as defense only. How does Adrian Beltre increase his OPS+ by 30-40% in hitters parks over SAFECO. Adrian Gonzlaez etc. there are drastic home away splits on offense and big splits in OPS+. The plus after OPS should adjust for park, but in a lot of cases it just doesn't give you the right information.

Jay Gibbons played RF for the O's from 2002-2007. In all but 2 seasons his UZR home splits were as good or better than his away splits. How did he manage it?

[Frobby]

Jay Gibbons played RF for the O's from 2002-2007. In all but 2 seasons his UZR home splits were as good or better than his away splits. How did he manage it?

To me, you just shot yourself in the foot. So, UZR says that Jay Gibbons was an above average RF at OPACY every year from 2002-05, but Nick Markakis had a negative UZR at home in both 2009 and 2010. Do you actually believe that Nick has ever had a year when he was worse defensively at OPACY than what we saw from Jay Gibbons when he was our starting RF? I don't, and that has absolutely nothing to do with Oriole bias, since they are both Orioles. You say my eyes fooled me, and Jay was in fact better than Nick at OPACY? Well, bullhockey. I know deep in my soul that Nick is way, way better than Gibbons ever was, and there is absolutely nothing you could ever say to convince me otherwise.

I'm done arguing with you about this. It isn't an argument where I can "prove" myself right, and I am not going to convince you, and you aren't going to convince me.

[El Gordo]

To me, you just shot yourself in the foot. So, UZR says that Jay Gibbons was an above average RF at OPACY every year from 2002-05, but Nick Markakis had a negative UZR at home in both 2009 and 2010. Do you actually believe that Nick has ever had a year when he was worse defensively at OPACY than what we saw from Jay Gibbons when he was our starting RF? I don't, and that has absolutely nothing to do with Oriole bias, since they are both Orioles. You say my eyes fooled me, and Jay was in fact better than Nick at OPACY? Well, bullhockey. I know deep in my soul that Nick is way, way better than Gibbons ever was, and there is absolutely nothing you could ever say to convince me otherwise.

I'm done arguing with you about this. It isn't an argument where I can "prove" myself right, and I am not going to convince you, and you aren't going to convince me.

It's not Orioles bias, it's Markakis bais. The fact is UZR which you think is skewed against Nick in RF, does not show similar results for Jay Gibbins. Also Matos, Matthews, and Patterson don't have the same level of trouble in CF as Jones does. We cant know for sure until we can see the numbers for opposing players in OPACY, but I suspect the problem is not the park effects. And I like Nick too.

[Frobby]

It's not Orioles bias, it's Markakis bais. The fact is UZR which you think is skewed against Nick in RF, does not show similar results for Jay Gibbins. Also Matos, Matthews, and Patterson don't have the same level of trouble in CF as Jones does. We cant know for sure until we can see the numbers for opposing players in OPACY, but I suspect the problem is not the park effects. And I like Nick too.

Well, Nick is my favorite player, no bones about it. But there still isn't a doubt in my mind that he's a much better RF than Jay Gibbons was on his best day. And I was never one to bash Gibbons' defense, either, unlike a lot of others. By the way, in Tango's 2004 and 2005 polls, Gibbons scored in the 30's, while Nick consistently scores in the 70's. So, it's not just me.

[CA-ORIOLE]

To me, you just shot yourself in the foot. So, UZR says that Jay Gibbons was an above average RF at OPACY every year from 2002-05, but Nick Markakis had a negative UZR at home in both 2009 and 2010. Do you actually believe that Nick has ever had a year when he was worse defensively at OPACY than what we saw from Jay Gibbons when he was our starting RF?

Take a look at Gibbons best year in 2004 and the ridiculously bad RF's (Kevin Millar/Miguel Cabrera/Matt Stairs etc.) of that year. Compare them to Nick's competition this year in RF. There is your answer. The stats are relative to that year and defense is just better now in RF (and probably a lot of other positions) than it was back then.

[El Gordo]

Well, Nick is my favorite player, no bones about it. But there still isn't a doubt in my mind that he's a much better RF than Jay Gibbons was on his best day. And I was never one to bash Gibbons' defense, either, unlike a lot of others. By the way, in Tango's 2004 and 2005 polls, Gibbons scored in the 30's, while Nick consistently scores in the 70's. So, it's not just me.

The major difference between Gibbins and Markakis is Nick's arm. Nick gets a lot of points for his ability to hold runners, but his range sadly is not much better than Gibbons. Deal with it.