clock menu more-arrow no yes mobile

Filed under:

Barton Would’ve Had It - A Criticism of UZR

Organic, free-rangey Barton.
Organic, free-rangey Barton.

If you don't already listen, ESPN's Baseball Today podcast is one of the few good things that come out of ESPN these days. It's worth a listen if you have 30-40 minutes in your day where you're performing a mindless task that leaves your ears open.

On Friday's episode, one of the cavalcade of co-hosts, Mark Simon, mentioned that Brett Lawrie has skewed defensive stats. Depending on your choice of defensive metric, Lawrie already has saved somewhere between 5-7 runs in less than two months worth of defensive chances. For the uninitiated, UZR is one flavor of defensive metrics that exist that attempts to rate fielders at each position against each other. So, all the first basemen are rated against each other based on the likelihood that the average guy at that position that year would have made the play. Positive numbers mean that a fielder has saved runs, whereas negative numbers mean that a fielder has given up runs. Plus or minus 10 in any direction means that fielder is pretty good; anything above that puts someone in rarefied air. Think Andruw Jones in his prime in CF for someone really good, and Jack Cust in LF for someone really bad.

At any rate, Simon delved into data with help from ESPN's research department and found that one reason for Lawrie's early success was a number of plays he had made near second base due to an aggressive shifting program that the Jays and many other teams have begun to employ against dead-pull hitters. In other words, because of where a normal 3B would have stood relative to where Lawrie was actually standing before the play, he is credited by UZR with having great range, when in actuality he was simply best positioned to make that play because of the shift. This is where UZR begins to break down - at one point is someone simply a good defender, versus someone whose team got lucky a few times and guessed right on positioning.

How does this affect Barton? Take the play he made last night, ranging behind him and to his left to snag the popup off of Mike Trout's bat to end the 3rd inning. A good play, to be sure, but that seemed to be a play better suited for Jemile Weeks, who came from farther away, but had a superior angle. Barton also makes this play well In Oakland, where a fielder can range far to his left to snag foul pop-ups. Finally, a first basemen can also pick balls to his right, but this time, impeding the ball from getting to the 2nd basemen, whose momentum is taking him toward the direction has to throw to make the out.

If a player makes enough extraordinarily range-y plays, he ends up being credited where perhaps this is not a measure of his true talent. Worse, if a range-y play is actually due to superior positioning, it can paper over a fielder not making the routine plays that 90% of fielders should indeed make. Is this why Barton's 2010 UZR of +10 was so good? Did he make a couple extraordinary plays but then muff some easy ones? It is commonly cited that UZR needs approximately 3 seasons of data to be useful, but let's be real here: analysis is mostly season-based.

So, what kind of fielder is Daric Barton? Is 3 seasons even to rate him, knowing that UZR is such a flawed metric? Is he the best guy the A's can find when taken as a whole? Discuss, if you wish.


Join baseballgirl for today's game thread, and me tomorrow against the Rangers. The A's will try and cleanse the slegnA away using a Colonic in this short series in front of me, the 500 or so other transplanted A's fans who show up to the Big A, and 29,800 of my worst red-clad enemies. 4 PM start... WTF MLB?