So after the Yankees game today I came across this Newsday article on Judge's high praise for Urshela and how much he believes Ushela has contributed to the team this year and it got me curious enough to wade into the miasma of baseball defensive statistics (Well, it seems that way to me). I've tried looking at it before but I can't make much sense of it. It seems like there are stats on top of stats intended to try to make sense of, or better quantify, some underlying stat that never worked well, to begin with.
So here I am expecting to some great stat illuminating Urshela's dazzling defense at third base, but instead, I find some nonsense trying to say that his defense has actually cost 4 more runs against than the average at third base. That is complete CRAP! He's a human highlight reel waiting to happen, seemingly just about every game. He gets to balls and makes plays where 75% of the rest of third basemen aren't even getting to the ball, let alone recording an out as well.
These defensive stats seem like garbage, or am I missing something? I also wanted to get a better read on his defense versus Andujar's but what I can make of those defensive statistical measures is of no use. On that question, I would not be surprised if Urshela accounted for 40 + outs more than Andujar. When you think about all the balls that Urshela gets to that Andujar isn't, plus his sloppy glove work and slow play getting the ball to first, I think it could be even more than 40 runners difference.
Curious to get people's thoughts on defense stats, as well as Urshela's contribution.
Yeah, he's got some errors but I think he's made up for much if not all of them with those sparklers.
His error rate is really very high. Looking even just at the guys with the most errors in the league, he rates quite poorly when it comes to errors.
Tim Anderson 582.2 innings 291 chances 16 errors
Jose Ramirez 747.1 innings 250 chances 15 errors
Jonathan Villar 771.1 innings 368 chances 15 errors
Gio 558.2 innings 184 chances 13 errors
Tim Beckham 515.2 innings 264 chances 12 errors
Other Yankees
DJL 722.2 innings 361 chances 5 errors
Gleyber 724.2 innings 327 chances 11 errors
Andjuar 2018 1169.1 innings 286 chances 15 errors
Gio's errors are a pretty material drawback to his defense. One can certainly disagree on the relative weightings of errors vs strong defensive plays in a total defensive value measure, but it's not surprising to me that he grades out somewhat negatively given the high frequency of errors.
IDK, but the Yankee IF defense seems a lot better than last year.
Quote:
Yeah, he's got some errors but I think he's made up for much if not all of them with those sparklers.
His error rate is really very high. Looking even just at the guys with the most errors in the league, he rates quite poorly when it comes to errors.
Tim Anderson 582.2 innings 291 chances 16 errors
Jose Ramirez 747.1 innings 250 chances 15 errors
Jonathan Villar 771.1 innings 368 chances 15 errors
Gio 558.2 innings 184 chances 13 errors
Tim Beckham 515.2 innings 264 chances 12 errors
Other Yankees
DJL 722.2 innings 361 chances 5 errors
Gleyber 724.2 innings 327 chances 11 errors
Andjuar 2018 1169.1 innings 286 chances 15 errors
Gio's errors are a pretty material drawback to his defense. One can certainly disagree on the relative weightings of errors vs strong defensive plays in a total defensive value measure, but it's not surprising to me that he grades out somewhat negatively given the high frequency of errors.
Errors are not really how you judge defense. No more than using RBIs tell how productive a hitter is. This isn't the 1980s where people just looked at errors and that was it for fielding. Range, positioning, etc are huge factors. Sometimes players make errors cause they get to balls other players simply don't have the range or quickness to get to.
I want you to tell me you think Gio Urshela is a bad defensive 3B with a straight face cause he has made some errors. You clearly aren't watching the Yankees every day if you thinks so. Again, this isn't the 1980s where you can look at simple stats like wins and losses for pitchers or batting average for hitters. Cause watching him play 3B every day he is easily one of the better fielding 3B in the league and has made numerous spectacular plays. The Yankees aren't going out and looking for high strike out pitchers like they were before the season cause they know the right side of their IF is so good now with Didi, DJH, and Urshela.
Quote:
In comment 14498815 Dave in Buffalo said:
Quote:
Yeah, he's got some errors but I think he's made up for much if not all of them with those sparklers.
His error rate is really very high. Looking even just at the guys with the most errors in the league, he rates quite poorly when it comes to errors.
Tim Anderson 582.2 innings 291 chances 16 errors
Jose Ramirez 747.1 innings 250 chances 15 errors
Jonathan Villar 771.1 innings 368 chances 15 errors
Gio 558.2 innings 184 chances 13 errors
Tim Beckham 515.2 innings 264 chances 12 errors
Other Yankees
DJL 722.2 innings 361 chances 5 errors
Gleyber 724.2 innings 327 chances 11 errors
Andjuar 2018 1169.1 innings 286 chances 15 errors
Gio's errors are a pretty material drawback to his defense. One can certainly disagree on the relative weightings of errors vs strong defensive plays in a total defensive value measure, but it's not surprising to me that he grades out somewhat negatively given the high frequency of errors.
Errors are not really how you judge defense. No more than using RBIs tell how productive a hitter is. This isn't the 1980s where people just looked at errors and that was it for fielding. Range, positioning, etc are huge factors. Sometimes players make errors cause they get to balls other players simply don't have the range or quickness to get to.
I want you to tell me you think Gio Urshela is a bad defensive 3B with a straight face cause he has made some errors. You clearly aren't watching the Yankees every day if you thinks so. Again, this isn't the 1980s where you can look at simple stats like wins and losses for pitchers or batting average for hitters. Cause watching him play 3B every day he is easily one of the better fielding 3B in the league and has made numerous spectacular plays. The Yankees aren't going out and looking for high strike out pitchers like they were before the season cause they know the right side of their IF is so good now with Didi, DJH, and Urshela.
And just to add to your point. A bunch of those errors came off of balls that were just ridiculous and the error came from not quite being able to complete a hurried throw from some crazy off balance body position. I remember some of those in the beginning of the year. Thinkning it wasn't quite fair to give an error. The other thing that i'd been thinking about is that we lose some of his performance with a first baseman who can't pick his throws very well. With a good fielding 1b Gio is several errors fewer.
There was one where he ranged way over to the line and the throw pulled 1B off the line. This is where the stats don't tell the story. Had he never attempted the throw in the first place - it wouldn't have been an error.
With Urshela, it seems like he's being penalized due to the high rate of errors. However, several of those are likely balls other guys wouldn't even touch and those wouldn't be dinged for. So with the smaller sample size, better to trust your eyes and general reputation than a stat (that even statheads would say is unreliable at best in smaller samples).
Urshela made some routine errors, more so than perhaps his peers. In run-value context, the negative on an error of a routine play is is as costly as what is gained from a phenomenal defensive save, even if our brains dismiss the error as "shit happens" and savor the sparkling defensive play as something more than just one play-- because it was so magnificent and unexpected.
To echo an earlier point, advanced fielding metrics need a lot of data before they become sustainable or meaningful. I've read (maybe it has changed in recent years), that it takes 3 years worth of fielding data to have meaningful results-- to have the stability of the results of one offensive season. Think about that-- think about how irregular an entire season the plate can be with regard to the true talent of a player.
By the end of the season, the fielding data will be a little better, but there is still a ton of variance and I wouldn't read too much into it in such a small sample.
Link - ( New Window )
I do think it's possible our perception is a bit skewed by seeing him make plays at the edge of his glove (unable to gauge his speed), because he is far, far better than Andujar, because he came out of nowhere, and because he has been such "House money" at the plate.
There was one where he ranged way over to the line and the throw pulled 1B off the line. This is where the stats don't tell the story. Had he never attempted the throw in the first place - it wouldn't have been an error.
Exactly
I do think it's possible our perception is a bit skewed by seeing him make plays at the edge of his glove (unable to gauge his speed), because he is far, far better than Andujar, because he came out of nowhere, and because he has been such "House money" at the plate.
So, for the benefit of speaking from a constructive perspective, I just watched a few of his highlights videos and here's what I see. He's eyeing up the play as it unfolds. I don't see slow of foot. I see him reading the bounces, speed, and spin on the ball getting himself in the best position to be able to grab it and then make a throw from a familiar, off-center and falling position with enough time to make the play at first. He's processing all that in a very short period of time. These plays dazzle (and they do dazzle, not just appear to)because he consistently excels at reading difficult balls, e.g. the short hop, while being able to get to the ball, collect it and throw it smoothly from an otherwise awkward position, with strength, accuracy and timing to just make the play at first.
He makes these very difficult plays consistently where others don't get to the ball and if they do aren't as likely to catch it and then make the play at first. I don't think he's slow footed. Looks to me like he's got an awesome ability to read the play right off the bat and the athletics to pull it all off.
There were a few plays where he had to dive for a screaming liner and it looked like he reacted quick enough. Maybe not cat-like reflexes of some others, but well enough. I think his sparklers, that average third basemen aren't getting, far exceed the number of errors on the basic plays. I think he's earned his accolades.
Doesn’t meant we can’t try to get some numerical sense but it’s like and mathematical model. Who is going to decide what goes into that model? At some point its just over analysis. Like you need some stat model
To tell you how and what time to take a shit.
Kid has been a Godsend this year both in the field and even at the plate. Anyone telling you his D Isn’t helping l is nuts
Doesn’t meant we can’t try to get some numerical sense but it’s like and mathematical model. Who is going to decide what goes into that model? At some point its just over analysis. Like you need some stat model
To tell you how and what time to take a shit.
Kid has been a Godsend this year both in the field and even at the plate. Anyone telling you his D Isn’t helping l is nuts
It’s hard to quantify intangibles. So when you talk in terms of “clutch” or “ having the ‘it’ factor” stat guys can’t point to something that seen by watching a guy vs numbers
However, there are also other stats that fail to meet the same standard. Many around fielding are that way. However, stat enthusiasts often look at all stats as being worthy of telling the story and they often put equal weight in them.
Hell, football is nowhere near the proficiency in breaking things down like baseball does with advanced stats, but that doesn't stop multiple references to PFF each week here.
However, there are also other stats that fail to meet the same standard. Many around fielding are that way. However, stat enthusiasts often look at all stats as being worthy of telling the story and they often put equal weight in them.
Hell, football is nowhere near the proficiency in breaking things down like baseball does with advanced stats, but that doesn't stop multiple references to PFF each week here.
I think you make a good point, especially in the comparison to some of the advanced player metrics in football such as those provided by PFF (which is not to say that there are no viable advanced metrics in football; there absolutely are, particularly some of the NFL's own "next gen" stats).
Defensive proficiency in baseball combines two things that an unwitting observer cannot accurately quantify: what was the players assignment at the time (typically thought of in terms of positioning), and then being able to sort through the noise of what appears to be an amazing play, visually.
In the case of the former, if you don't know what the defensive positioning was (or if you don't account for it), you're grading on a curve without even being aware of it.
With the latter, as we're seeing with Urshela, you can simultaneously be a very good fielder and still sometimes be a defensive liability. A runner who reaches on a throwing error by GU isn't any less costly or dangerous to your chances to win than a runner who reaches on a hit on the same ball to a lesser infielder.
Whenever a debate like this crops up around defensive prowess in baseball, I'm reminded of the way many fans (and media) drooled over Jim Edmonds play in CF, while Bernie Williams often seemed underappreciated. And the only reason, IMO, was that Edmonds dove for fly balls all the time whereas Bernie had a great jump and first step and often tracked balls down without having to dive.
And that's not to say that Edmonds was not a very good or even great CF; he was. But when he booted a ball on a dive attempt, it went for extra bases. Fans of his ignored that. We might be edging toward that territory with Urshela. He makes some great plays that suggest that he's a great fielder, or has the potential to be one. But that doesn't mean that we should ignore the fact that his error rate currently does provide a counterbalance to his superior range.