Mark Teixeira And Defense Metrics

RLYW continued their series on projections by looking at Mark Teixeira. While they found that he should have an excellent season at the plate, the defensive numbers on Tex were a bit surprising to me. They show that Mark is at best a slightly above average defensive first baseman:

One of the things that Teixeira also supposedly brings to the Yankees is a great glove. However, the statistics don’t necessarily agree with the scouting reports on that….

That doesn’t really look like the supposedly great glove we keep hearing about, although it’s worth noting that ZR and UZR don’t capture a 1B’s ability to scoop bad throws or to chase down foul popups, which are probably worth a few runs. He should at least be solidly average at 1B.

The scouting report on Teixeira is that of a Gold Glove caliber first baseman, and this incongruence between the numbers and the scouts brings me to my point. I love statistics, and use them liberally when it comes to hitting. However, in regard to defense, they are best used in conjunction with scouting reports, with neither element holding greater weight. Defensive metrics are surprisingly subjective, as they often include judgments made by an official scorer. Because they are a very inexact science, treating them as definitive or significantly better than scouting reports seems a bit silly. To some degree, we have to trust our eyes to properly place Teixeira on the spectrum of defensive ability.

Does this seem a bit out of touch with reality, maybe displaying an unwillingness to hear that Tex is not amazing defensively? Well, it seems that some baseball insiders with a statistical and analytical background agree with my reservations about defensive metrics. Here is a snippet from an interview that BP did with Shiraz Rehman, the Director of Baseball Operations and a statistical guru for the Diamondbacks:

I think that defensive metrics, on the aggregate, have improved significantly in recent times, thanks to the work of John Dewan, Bill James, and others. But in fairness, they are probably still imperfect, and as a result we still rely a fair bit on some of our internal assessments and subjective valuations in that area to come up with a comprehensive look at defense.

I totally agree. In regard to defense, the numbers often lie, and just pointing to them without seeing the player field his position would be irresponsible. I know that Derek Jeter is bad defensively because I can easily observe his lack of range, and the numbers confirm that. On the other hand, I find it hard to believe that Brett Gardner is the second coming of Willie Mays given his poor routes to balls, and I think that the defensive metrics probably overrate him a bit. Ultimately, it is important to find some sort of balance between the scouting reports and the metrics. Otherwise, there is no way that you can form an accurate conception of a player’s defensive ability.

Related posts:

1. Teixeira's defense a complex discussion
2. Mark Teixeira: The Swing Vote
3. Defense is the New Moneyball?
4. Mark Teixeira is nowhere to be found
5. Mark Teixeira thinks he's boring