20 June 2012
We need to be very careful whenever we discuss advanced defensive metrics, such as UZR (click here for a UZR explanation), when the sample size is small. Conventional wisdom says three years of defensive data is as reliable as one year of offensive data (due to factors such as number of fielding opportunities and the limitations in available batted ball data), so our current 67 game sample on Miguel Cabrera would be akin to 22 games worth of batting numbers.*
So we obviously can’t make any “true talent level” claims based on this small data set, but we also can’t say that UZR tells us what has happened so far this year. All we can discuss is what UZR has recorded as having happened so far this year. The distinction may seem small, but it’s important.
*We know that 22 games of batting data is full of noise due to random variation. For example, Ramon Santiago hit .163/.208/.224 through his first 22 games this season. Since then (29 games), he’s hit .284/.392/.388. Are either of those stretches his true talent level? No. In fact, we probably get closest if we just average the whole year together to get his .233/.318/.319 line.
But it wouldn’t be fun to simply put aside the data while calling it unreliable would it? Let’s take a look at his UZR data (as reported by FanGraphs) for this season.
As you can see in the table, UZR records Miguel Cabrera as having cost the Tigers four runs defensively compared to an average-fielding third baseman. The UZR/150 column tells us that this rate, if continued out for a 150 game season, would result in a net loss of 10.3 runs. That’s fairly bad to be sure (he’s currently last among qualified third basemen), but it’s not nearly as bad as many of us feared he would be.
I did some rough calculations before the season began, and found that Cabrera would need to be nearly 27 runs below average at third base in order for the experiment to be a complete failure. I estimated that Cabrera would end the season between -15 runs and -18 runs defensively (and I may have been one of the more optimistic prognosticators).
But to me, it’s more interesting for us now to look away from the totals to see how they are derived. The three columns on the left tell us double play runs (DPR), range runs (RngR), and error runs (ErrR). We can see that Cabrera is clearly limited by his range. We all expected this. He’s a big man, and he’s simply not able to move fast enough to get to a lot of baseballs, so it should surprise no one to see a big -7.20 recorded in that column.
Our worst fear, though, was that he was going to have poor range and mishandle playable balls and throw the ball away with regularity (this is what happened for the short time he played third for the Tigers in 2008). But that doesn’t appear to be the case right now. In fact, our UZR table displays Miguel as being above average in starting double plays and avoiding errors, to the tune of +3.20 runs.
I would call this a pleasant surprise so far (caveats still remain that the numbers can change rapidly), but it’s not entirely shocking. Even while at first base the past few years, Cabrera showed us that he still had a strong arm capable of making accurate throws. He seemed to be as good as anyone at starting the 3-6-3 (or 3-6-1) double play, and he was never afraid to throw to third to gun down the lead runner trying to advance from second on a ground ball to the right side of the infield.
We’re obviously a long way from declaring success or failure (although these numbers and the eye test suggest he’s not drowning), but it’s encouraging to see some of these metrics. We conceded that his range would be lacking well before the season began, but he’s otherwise displaying a good feel for the position.
Matt Snyder is the editor of The Tigers Den. He can be reached on Twitter @snyder_matthew.