I love my job, so very little about it frustrates me, but for one little thing: at times like this, I am often criticized for positions I did not take. For example, among the responses to yesterday’s Gold Glove reaction, we find reader Ryan:
Do you save time of watching the games and rely just on stats? You are crazy if you think Ellis or Hudson are more deserving than Cano. It seems as if you let the stats control your entire thought process on every aspect of the game.
(1) I didn’t say that Ellis or Hudson are more deserving than Cano. I said that the advanced metrics say that those two were better. Nowhere did I endorse what they say, I merely reported it. Further, I said, "As for those metrics’ opinions on Cano and Teixeira, they will tell you that they weren’t necessarily the best choices… but they weren’t poor choices either… These are minor caveats given that Teixeira and Cano deserve to be part of the conversation by any measure, whereas Jeter does not." That is hardly calling for the Cano result to be overturned. I personally am satisfied with Cano’s win. "He has worked hard to improve, and the results show on the field. Not only are the careless errors gone, but he sometimes makes the spectacular play." Whaddya want me to do, genuflect?
(2) Whereas I don’t let statistics do my thinking for me, I do appreciate them because they give me a good way to keep track of hundreds of discrete events that no human mind has the capacity to remember with clarity, like every ball that Cano swung at this year, or every one he fielded. They also allow me to be in two places at once. Because my focus is this column, I am a sedulous watcher of every play of every Yankees game, and as for tracking the rest of the league, I watch what I can in the remaining time available—entire games when possible, fractions of many more, and everyone’s highlights. Fortunately, there are two additional resources I can utilize: (A) there are other people whose job it is to watch 162 games of the A’s or the Twins and the rest, and I can read what they have to say about Ellis or Hudson, or question those writers directly, and (B) I have the stats, an objective record of what happened on the field. Throw in video and my notes from various games throughout the year and I can form a reality-based picture of most players despite not seeing every chance they got in the field. Now you know how I would have felt comfortable making an argument in favor of Ellis or Hudson over Cano... which I did not do.
(3) As long as we’re talking about objective records and stats, let us, as Dick Nixon used to say, make one thing perfectly clear: all statistics are is a record of what happened on the field. They require context to be understood, but for the most part they’re just a time-saving device. They can’t control your thinking anymore than a grocery list can. Again, no one’s memory is so good they can remember every hit or pitch, and it is much easier to be able to check the stats and see that Alex Rodriguez hit 30 home runs than it is to try to remember by tying bits of string around your fingers and toes for each shot he hit.
Fielding statistics add a level of complexity beyond mere memory augmentation; because the basic fielding statistics (assists, putouts, errors) are so inadequate, there is an act of interpretation that has to take place if you’re going to derive meaning from them. Because there are so many different approaches to making that interpretation, I don’t like to look at any one measure, but survey several and develop a consensus view of a player. The goal is not to get an exact number of runs that a player harmed or benefitted his team with his glove, but a sense of the degree to which he helped or hindered. As Colin Wyers showed in the piece I linked earlier, the consensus view of Jeter varies from mildly negative to morbidly pessimistic. In essence, there is no way to look at what Jeter has done in the field and come to a positive view of his defense.
Again, those statistics should be used to augment what you see with your eyes. My eyes tell me that Jeter makes relatively few errors but also gets to relatively few balls. That is not "the stats control[ling[my] entire thought process," but using the statistics as a quality check on my own flawed human perceptions.
That brings us to reader Rafael. I usually correct for spelling and punctuation, but I don’t know where to begin here, so we’ll just go with the original:
when you talk about Cano back in the day committing too many errors you don’t think elvis andrus 16 clearly makes Jeter a better choice. .989 fielding percentage is nothing to laugh at. you have no clue what you are talking about. you can’t use hypotheticals such as the ridiculous argument about Jeter’s range when he obviously makes almost 99% of the plays that he has a chance to. give the man credit instead of trying to bring him down. does it get you anywhere by being negative?? no, it makes you seem petty and childish. the only truth is, Jeter had the HIGHEST fielding percentage among short stops and the lowest amount of errors!!! that is all you need to know!
Here is why fielding percentage is not a good measurement of a player’s defensive abilities. It tells you a little bit about the balls he did field, and what percentage of them he handled cleanly. They tell you not a damned thing about the balls he didn’t field. Let’s make up an imaginary shortstop. Call him Larry Sheffield. Larry Sheffield has a bad attitude, and one day and decides he’s just not going to field his position. Ten balls are hit to him that day. He picks up the first one and throws it to first for an out. When the other nine comes, he doesn’t make a move after them, just stands still and watches them go by into the outfield. You know what fielding percentage calls that? A perfect day, 1.000. Sheffield accepted one chance and he made the play. All the balls he didn’t get to are invisible as far as that particular statistic is concerned.
Jeter presents a similar problem, even though his attitude is 180 degrees from that of our disgruntled imaginary infielder. A hypothetical game, this time with the Captain: a ball is hit right at Jeter. He picks it up and throws to Mark Teixeira for the out. Four others come his way. He hustles. He dives. He does spin-‘n’-jumps. He does a fouetté rond de jambe en tournant, causing many in the crowd to spill their beers in astonishment. Yet, somehow those other balls scoot past him into the outfield. Of the four balls that Jeter failed to intercept before they turned into singles, two of them were placed in a way that not even Ozzie Smith would have gotten to them. The other two, however, would not only have been picked up by Smith, they would have been picked up by the average fielder. Those are the balls that we’re concerned with when we talk about how Jeter’s defense affects the Yankees. What does fielding percentage say about those balls? Not a thing. Since Jeter didn’t touch ‘em, they don’t exist. He accepted one chance, fielded it without making an error, and therefore had a perfect day. As John Foley said in the comments, "you can’t make an error on a ball you never get to."
Andrus made 16 errors to Jeter’s six. He also accepted 106 more chances than Jeter in 12.1 fewer innings. Again, ten more errors, but over 100 more plays. Now, some of that difference could be the result of pitcher strikeouts, how many left-handed or right-handed batters the Yankees and the Rangers faced, random variations in the distribution of grounders, and we have to look at those things if we’re really going to compare the two. But let’s just take your argument on its merits. Andrus fielding percentage was .976 and Jeter’s was .989. That’s a difference of .013, which is meaningless. Take nine errors away, nine plays out of Andrus’ 106-play advantage, and their fielding percentages would be even. Nine errors seems a small price to pay for those 100-plus extra stops.
One other note on fielding percentage: official scoring is a joke, and the difference between a fielder being charged with an error and a pitcher being charged with a hit has become entirely arbitrary. As such, errors are one of those statistics we have to take with a massive grain of salt.
Let’s follow Rafael with "Jesse from NJ." This one is a bit of a mystery:
It’s funny when the stats are used for certain arguments and not on others. It is clear you guys don’t watch the games, plus I thought this was the Yes Network and not some other teams network. It is also funny how they vote on the allstar selections, and don’t even think about telling me the fans vote the players in, yeah maybe 1 or 2. Get a clue guys and write about something else
A few points, Jesse from NJ:
(1) I made a point of not using statistics to describe Jeter’s defensive game. As I have been saying, if you just use your eyes you can see his lack of range quite easily. Had I used statistics to make the argument, it would have been very negative, because there is no statistical method of evaluating defense that sees Jeter as a quality defensive shortstop. It’s amazing how Jeter partisans aren’t satisfied with his going to the Hall of Fame and being one of the most beloved Yankees in the history of the club. He also has to be acknowledged as perfect in all phases of the game.
(2) You’re right. We don’t watch the games. We play out the schedule with Strat-O-Matic cards and then analyze the results here. Watching the actual games is too much work and it requires us to miss too many episodes of "Glee."
(3) It is the YES network. The great thing about the YES network is that the powers that be have enough respect for the fan that they don’t expect us to hand you pablum and brummagem. Isn’t that neat?
(4) You’ve figured out the All-Star Game conspiracy! It’s not the fans who vote the players in, it’s the Trilateral Commission.
Here are three quickies:
Mr. Goldman, I respect your opinion, but you are an idiot.—Joe
Joe, I respect your right, having failed to out-think or out-argue me, or even try, to resort to abuse.
I Don’t know where your head is it must be where the sun don shine, because they desere it ansd so dose Garner.—Lew
I am reminded of a line in an old Tom Reamy story: "He sure wished he knew what people were talking about, at least some of the time."
hey steve why dont you check errors made by shortstops and see who had the less bet 10 it was jeter with less—Ken
Bet 10 what? I can’t use anymore poultry. Have we got it yet that errors or the lack thereof don’t say anything about the balls a player didn’t reach? Any of us could field 1.000 in the majors as long as we didn’t try to make any plays. You could go out to short, put your hands over your eyes, and curl up in a fetal position and field 1.000. A dead man could field 1.000 at any position. Sure, nothing hit to his zone would be caught, but again (and again and again) they don’t charge errors on plays you don’t make.
Having spent far too much time on this entry already, I will close with "mike," who confuses me by arguing that fielding percentage tells little about a fielder’s ability, but then says, well, you’ll see:
Let’s not have any crazy talk about Jeters fielding using some statistical mumbo jumbo calculations. Fielding ability never was and never will be able to be judged by a mathematical system. Fielding average tells little about a fielder’s ability. Recent hall of famer Joe Gordon did not always have a great fielding average because he got to so many extra balls that no one else could get to. Although Jeters average is always very good. Just watch Jeter every day and marvel at his ability. Diving for balls does not make a fielder better just flashier. Judgment on the field is the most important thing for a fielder.
Let me work through this; last set of numbered responses for the day:
(1) Didn’t use statistical anything, mumbo-jumbo or otherwise. Read it again.
(2) "Fielding average never was and never will be… judged by a mathematical system." Not perfectly, no, but they can give us a heck of a clue. And if you don’t like those, there are observation-based systems like zone rating (in all of its forms) and John Dewan’s plus/minus system. Those are eye-based, not stats-based. They happen to be in broad agreement with the stats-based systems as far as Jeter goes, but as Simon and Garfunkel sang, "A man he hears what he wants to hear and disregards the rest." I would add, "And you can’t do anything about it."
(3) On Joe Gordon: sometimes errors are the result of good range. Excellent; finally something I can agree with.
(4) "Just watch Jeter every day and marvel at his ability." I do, or did until this year—his ability at the plate and on the bases. I also marvel at his ability to not get within hailing distance of balls hit anywhere to his left, as well as the ability of almost every opposing shortstop to reach more of the same.
(5) "Diving for balls does not make a fielder better, just flashier." Another point of agreement! Diving, at least in the outfield, can be a very bad idea, because you take what might have been a single if you didn’t catch it and risk turning it into a triple. On the infield, well, I suppose there’s too much of a bad thing, but I grew up watching Graig Nettles play third base, and his ability to dive after a ball was both flashy and brilliant. There are players who dive just to dive, Carney Lansford most infamously, and look great while letting everything go past them, but that aside, if an infielder dives after a ball, he might get it. If not, it becomes a single. If he doesn’t dive, it becomes a single. The risk factor just isn’t the same as in the outfield.
(6) "Judgment on the field is the most important thing for a fielder." I’m not sure what this means, or what the relevance is to an infielder. Also, judgment off the field is pretty important too; you don’t want your star shortstop playing in traffic after the game is over.
As for me, having written another 2500 words on this topic, I think playing in traffic is exactly what I’m going to do next.