It became pretty clear during the 2011 season that a great career was winding down. On August 25th, the Yankees hit three grand slams en route to a 22-9 drubbing of the Oakland A's and Jorge Posada was again relegated to the bench. But in a moment of fun during his swan song, Joe Girardi agreed to let Posada play second base for the ninth inning, a position that he hadn't played since the minor leagues.
Fittingly, the only ball hit in Posada's area was the twenty-seventh out and he made an adventure of it, firing a low throw that required a great play from first baseman Nick Swisher. But how would you evaluate Jorge Posada the second baseman? He had one defensive chance and made the play. The best second baseman in history could have done no better.
For a second example, imagine if the Yankees call up Doug Bernier and he hits the first pitch that he sees for a home run. He would have a 1.000 average on the season and -- at that point -- be rated as a flawless batsman. Which is a roundabout way of illustrating that statistics measure results and not ability.
Ability is a nebulous concept that requires insight. Given enough result-oriented data, we can draw a lot of conclusions about ability, but it does take time and it is an inexact process. Jorge Posada has a literally perfect statistical record at second base, but common sense screams that he would have very little ability to play the position effectively.
And that's what I think people lose sight of, especially when defense is measured. While Curtis Granderson might hit two home runs one day and strike out four times the next, his defense just pretty much is what it is. There are no glaringly obvious changes in his speed, hands, or arm strength from game to game, and the great plays or egregious gaffes tend to be rare. So the idea exists that his defensive rating should be a refelction of his ability to play the position, which is more or less constant.
To draw another parallel, think of batting average. It may not be representative of his true ability as a batter, but Andrew McCutchen has 160 hits in 462 AB, so he has a brilliant .346 average. McCutchen is a career .292 hitter and hit .259 last year. Even his biggest fans would probably admit that -- through a combination of a career year and some good fortune -- he's playing above his head right now. Those hits are tangible and no one can take them away, but what is being measured and recorded are events and not McCutchen himself.
People tend to be on board with that concept when it comes to offensive production. Guys have hot streaks, slumps, great years, and down years. They're different when they're young and raw and when they're old and declining. They can make adjustments in their approach, develop power, move to a more friendly ballpark, or any number of other things. There tends to be a healthy uncertainty that someone's offensive numbers define their ability over the course of only a season or two.
Defensive statistics work the same way, despite the fact that they are rarely viewed in that appropriate context. Especially outfield defense. The question being answered is have you caught the ball and not can you catch the ball.
And so when you see that Curtis Granderson's -16.2 UZR is the worst rating in all of baseball, it's important to keep in mind what that is telling you. Statistics do not have the capacity to measure Granderson's defensive ability, but they can illustrate the fact that he hasn't caught a lot of balls this year.
Granderson has played in center field for 1076 of the 1118 Yankee innings so far in 2012. Yankees pitchers have allowed 1041 catchable fly balls* -- which I scaled down to 1001 to compensate for playing time -- and Granderson has 267 put outs (26.6%). In all of baseball, 32173 catchable fly balls have been allowed and center fielders have 9364 put outs (29.1%).
*I'm defining a catchable fly ball as a fly ball that is not a home run. This does not include balls classified as line drives.
For his career, Granderson has 2380 put outs in 8018 innings, or 0.297 put outs per inning. This season he is at 0.248 put outs per inning.
Yankees pitching has allowed a double or a triple on 9.2% of the catchable fly balls that they have given up. The league average is 8.4%. Granderson is catching balls at a much lower rate than both league and career averages, and it is compounded by the fact that plays not made in the outfield have a much higher chance of turning into extra base hits.
There are plenty of issues that introduce uncertainty into that kind of analysis, mainly classifying balls in play as line drives, fly balls, and pop ups and determining which "should" be caught and by whom. But however you parse the numbers, Granderson just hasn't turned a lot of batted balls into outs and his defensive rating is a reflection of that.
There are a myriad of mitigating factors that could be the driving force behind that. I would bet that defensive positioning will be a favorite choice for a lot of you. Maybe he has seen a disproportionate number of tough chances or is having trouble picking up the ball. You often hear about the importance of hitters seeing the ball well and it follows logically that outfielders would require the same skill. Maybe he's lost some confidence or maybe he's a step slower or maybe we're just experiencing a golden age of center field defense that he is being compared to.
There's no great way to figure out exactly why Granderson's defense has rated so poorly this year and what exactly that means. Normally a statistical blip on the radar, especially when dealing with UZR, tends to be an overstatement and you can expect some regression to the mean. It is very unlikely that Granderson's true ability is as a -20 run center fielder, but that's what the results have been so far, as best we can measure them. It's an important distinction and one that is often missed. So that's why I wrote this post.