The Angels' outfield defense was supposed to be a many splendored thing. With Josh Hamilton supplanting Mark Trumbo in right field, Mike Trout shifting to left and Peter Bourjos returning to a starting role in center, the Halos had arguably the best defensive outfield in all of the land.
We even joked how the Halos could play just Trout and Bourjos and still have nary a ball drop in the outfield. Ha! Ha! Ha! LOLZ. /fart sounds
Intellectually, that all still makes sense. Mike Trout and, when healthy, Peter Bourjos have range on range on range. And that Josh Hamilton guy is no slouch either when playing in the corner rather than center where he just doesn't have the athleticism anymore. The problem is that our eyes thus far hath been deceived... or at least that is what the defensive metrics would have us believe. Just look at this table of how the Angels' preferred starting outfield grades out in the most popular advanced defensive metrics. SPOILER ALERT: Negative numbers are no bueno.
You get a negative UZR! You get a negative FRAA! You get a negative DRS!
What in fresh hell is going on here? The obvious caveat here is that advanced defensive metrics for roughly a quarter of a season are full of noise and equine excrement, but it strikes me as weird that all three of main defensive rating systems rate all three outfielders between below average and Delmon Young-esque.
Oh but the weirdness doesn't end there. Guess who has a UZR/150 of +12.7, DRS of +1 and FRAA of +0.1? Mark Trumbo because of course Mark Trumbo does. Nothing about this Angel season makes any damned sense, why should we expect the defensive metrics to be any different?
I'm not even going to try and dissect each metric to explain how it is that the Angels' trio is being underrated compared to what the highlight reels suggest. Defensive metrics are weird and I am not nearly well versed enough in them to explain away their fluctuations. For all I know the Angels are doing something weird with their outfield shifts. Maybe it is just some odd bit of bad luck. Maybe it is somehow the result of the Angel pitchers being terrible because everything is this crappy pitching staff's fault. Or maybe Trout and Bourjos and Hamilton really are having a rough stretch on the defensive front.
It is because of the weirdness of the defensive metrics that makes me think this is much ado about nothing. Trout is amazing and Bourjos is amaze balls. Both of them (and Hamilton too) are still making a ton of out of zone plays and still, when healthy, providing outfield defense porn on the regular. They'll be fine in the long run, but the metrics remain confusing in the short run. It is just another oddity of this already confounding season.
|Like MWAH on Facebook||Follow MWAH on Twitter|
I have a question. Humans made these statistics to better describe what they saw. However, how do we even know these work at all in describing defensive ability. Let me explain. OBP is OBP, BB is BB, HR is HR, these kind of stats are just counting and doing some easy math. How we interpret them is up to us but they do not change. No one can subjectively define what a walk or home run is. It is objective. However the defense stats seem to be the exact opposite. When these stats were created didn't the creators of each stat put more emphasize into whatever category he thought better described defensive ability, bringing more importance to certain abilities than others?
For example lets say I am creating a defense statistic. I am trying to calculate 3B effectiveness. From watching games I see, subjectively, that Adrian Beltre is an absolute beast of a defensive third basemen, but I don't really like Evan Longria's glove. Now I know for certain that I can with enough tinkering make my stat say that Beltre is better than Longoria, but the problem is just as subjectively, someone can make their stat say Longoria is better than Beltre. So the question is, and maybe you can enlighten me on this point, which stats would be correct and where is the proof? All these stats say his fielding is good, or his fielding is poor, but is there any concrete, certifiable evidence that shows a defensive metric is pure and unbiased? Because any stat that says Trumbo is a better fielder than Trout and Bourjos, that can't be right, can it?
@erstadfan17 Defensive metrics are at their foundation event based. They measure binary things. Either a guy caught the ball or he didn't. He made an error or he didn't. There isn't really any subjectivity to that. Where there is wiggle room is comparing a play to whether or not an "average" player would've made the play. My understanding is that UZR does this all via video and data calculation whereas DRS uses human scorers to make those determinations, so there is definitely a margin for error in that rating.
The Beltre-Longoria example though is not a real concern. People create theses metrics to get an objective measure, not to prove a point. Even if they did, there are enough people involved in gathering the data and tweaking formulas that any one person's bias would be removed, assuming there is a way for that bias to even influence the metric, which is not possible in most cases.
The real issue with defensive metrics is the data available to the public is somewhat insufficient. As a result, the general agreement is that the data is only really conclusive over long spans. Some say as long as three years. There is noise in the data which is why we see what we see now with the Angels over 44 games. It is just a funny thing and probably doesn't mean anything. I'm sure it will all even out by the end of the season.