UZR Tidbits Through May 17th
Almost a fourth of the way into the season, let’s take some position-by-position glances at the best and worst defenders.
1B
Best: Chris Davis (3.9), Ryan Howard (3.2), Lyle Overbay (2.9)
Worst: Jason Giambi (-3.9), Nick Johnson (-3.4), Joey Votto (-3.3)
2B
Best: Ian Kinsler (5.4), Rickie Weeks (5.4), Brandon Phillips (5.3)
Worst: Skip Schumaker (-7), Dan Uggla (-5.7), Chris Getz (-5)
3B
Best: Ryan Zimmerman (6.7), Joe Crede (6.4), Evan Longoria (5.6)
Worst: Michael Young (-7), Josh Fields (-4.1), Chipper Jones (-3.8)
SS
Best: Marco Scutaro (4.9), Elvis Andrus (4.2), Ryan Theriot (3.6)
Worst: Yuniesky Betancourt (-8.2), Khalil Greene (-4.4), Miguel Tejada (-4.3)
COF
Best: Nyjer Morgan (9.9), Jay Bruce (8), Brandon Moss (7.7)
Worst: Jason Bay (-9), Andre Ethier (-8.4), Adam Dunn (-8.2)
CF
Best: Mike Cameron (7.8), Matt Kemp (7.3), Franklin Gutierrez (5.7)
Worst: Shane Victorino (-7.5), Vernon Wells (-5.8), Elijah Dukes (-3.8)
A few other tidbits:
Proof that this amount of UZR data is pretty useless in predictive value, Carlos Beltran ranks as the fourth worst center fielder in the entire league at -3.5 runs. Over the last three years, Beltran’s UZR have been 8.8, 1.2, and 5.3. Even if you think his skills have declined, he’s unlikely to keep up this pace, which would have him at nearly -12 over 150 games.
The Pirates are your new team UZR leaders at 17.5, ahead of the Rays (16.3), Reds (15.2), Rangers (15.2), and Brewers (14.5). Meanwhile, on the other side of the spectrum, the Nationals (-18.3), Mets (-14.4), White Sox (-14.4), Red Sox (-12), and Orioles (-9) rate as the worst set of gloves in the league.
Are these numbers already park adjusted? I ask because of Jason Bay’s ranking. He looks terrible in the field to me either way.
More than not being park adjusted, they’re subject to extreme fluctuations in limited samples. So, while it’s fun to look at these numbers every week, at this point, you should take them with a big grain of salt. As the post points out, there’s very little predictive value in these numbers so far. Most players will be closer to average (i.e. 0) at the end of the season than they are now.
UZR is park adjusted.
And Jason Bay since 2007 has always been considered awful by UZR whether in Boston or not.
It does seem remarkable that the Red Sox traded the old worst LFer in MLB for the new worst LFer in MLB.
How good is the park adjustment for CitiField and the new Yankee Stadium? Problems with the former might have something to do with Beltran’s low fielding numbers.
Im writing this assuming UZR is position dependent.
Whats the consensus of using UZR for SS/CF bc those are the “premier” defensive positions? Since most SS/CF players are going to be very skilled so comparing them to the “average” seems to be a bad method of evaluation.
Think of it this way, would you make a “UZR type” evaluation for the IQ levels of MENSA members. Since the net sum of UZR should be close to, if not exactly 0, inherently there need to be ppl with negative values, and every MENSA members has an IQ above the 98th percentile.
Yes, UZR is position dependent and it is important to remember that a +1 SS is not the same as a +1 RF.
This is taken care of when assessing a player’s overall value in the positional adjustments in our win value section where players get credit for playing more skilled positions. When looking at UZR numbers independently, it’s just something you have to keep in mind.
If you’re trying to decide which among the MENSA members is best at a certain task, you still want some way of comparing them against each other. An 0 CF is a much better defensive player than a 0 1B, but there’s no reason to suppose he’s not. They’re just not meant to be compared against each other across positions.
On an unrelated note, shouldn’t Span be ahead of Gutierrez in LF? He’s at 5.8 right now, despite (a) having started only 12 games there and (b) ineptly playing a bloop single into an inside-the-park HR the other day…
I think part of my “problem” with the UZR is that it assigns negative values to SS/CF and theres a natural perception of someone being bad when they have a negative value.
Also, since the UZR are “year” dependent, how consistent are the ratings from year to year? Are there noticeable fluctuations from year to year in the “average players UZR?” Theoretically a player could play the same exact defense but have 2 different UZR ratings bc of how the other players played those years.
So can the fact that Victorino, Beltran and Chipper are rated as below average defenders finally make people stop quoting UZR after 1 month?
It’s a fun exercise. Try to not get worked up about it.
And maybe Beltran hasn’t played OF well so far this season? I don’t watch Mets’ games, so I couldn’t tell you.
Chipper Jones is a terrible defender, so UZR got that right.
For what it’s worth, Dewans plus/minus also suggests Beltran is off to a slow defensive start in ’09. It also thinks Victorino, Wells and Dukes are near the bottom.
“Yu-ni! Yu-ni! Yu-ni!” Wish I could blame the small sample size, but Yunieski really is this bad. It’s the strangest thing – his feet move but he doesn’t go anywhere.
I watch Victorino on a daily basis, and have a hard time seeing Victorino as the worst CF in baseball. He may not be the best, but certainly not the worst.
In the real world, those making measurements estimate uncertainty. So if you say based on X samples tested, the average measurement Y defines the property at the 95% level, thats fine. Another way is to say Lugo is -48 UZR/150 +/- 40, a large uncertainty due to SSS (just throwing it out there).
I still look at UZR/150 and plus/minus even with SSS, especially of the number is extremely positive or negative, and use it if I have observed the player enough and have confirmed it with my eyes. Lugo for example has lost range due to his surgery, and the stats reflect what I have seen. he may not be -48 at season end, he might recover some range, but is likely to be negative in the end.
To say Beltran is a worse fielder going from 5.2 to -3.5 (1/4 season) is not an indication of the stats being useless, it is ignorance of the uncertainty due to SSS. Unfortunately, the stats publish the uncertainty, so I guess we can say in Beltrans case it is at least +/- 8.7. Players fielding does decline with age and with injury, so waiting until the season ends to use the stats is not an option for teams or fans. Just so long as they are not taken as gospel.
Even with a full season, or a 3 year average, is the stat accurate enough to state that an average UZR if 5.3 is better than 5.1 conclusively, or with 95% or 99% confidence? The observers who record the data certainly do not do so the same way or with the same accuracy, and unless they are assigned randomly to games, which seems unlikely, biases are introduced. Then there is the accuracy of the observation and measurement, even for one who is very consistent, since much of it is visual and subjective (how hard it is hit for example, liner or fliner, etc).
Manny’s sudden improvement defensively at 37, and the younger Bay significantly declined. His worst year when he played a full season at Pittsburg was 2007 at 11.3, likely due to a bad knee that year. Then it went to 18.3 playing 1/2 the season with the Red Sox, and now playing full time with Boston it is 21.3.
Manny in 2007 was -28.3 in a full season in Boston, -3.2 is a season half in LAD, and -13.9 this year. So call it -6 over 1 1/4 season (more than 1/2 in LA), and this year he was bothered with a hamstring and a shortened ST.
Bays biggest issues defensively are his arm accuracy, speed of release, and picking up GB quickly. UZR does not measure this. His range is fine, although he tends to play too deep and balls Manny gets because of positioning he does not get, so is not only losing points on the wall, but on balls in front of him.
Yes, there must be something to that UZR after a month and a half. Unless Ryan Howard is working really hard on his defense.
Howard dropped 20 pounds in the offseason and does look a little quicker in the field. His throwing also appears to have progressed to “below average” from “cover-your-eyes terrible”. It wouldn’t surprise me at all if his UZR is improved this year. I doubt he’ll finish the year as the second fielding first baseman in the game, but he may well be decently above average. He was always perceived as a worse fielder than he really is anyway.
He is. Or, he spent the off-seaon/spring training working with someone (help me out here) on it quite a but actually. The little I’ve seen of Howard he certainly looks more nimble and better positioned out there.
Sam Perlozzo (third base coach)
Howard’s drastically increased UZR is not a random fluctuation, his fielding this year has actually been above average. Victorino’s UZR is puzzling to me because his range is good. I wonder if it has anything to do with Werth’s range and the frequency that Utley and Rollins tend to take popups in shallow center.
Thanks for the Information, thanks for your useful Article. Also check these nice Blog. estate agents belfast