Designated HitterDecember 01, 2006
The Worst Minor League Defenders
By Jeff Sackmann

When a prospect gets a reputation as a bad fielder, his chances of future success go down, fast. An infielder like Joel Guzman could be permanently moved to a less defensively-demanding position such as first base or a corner outfield spot. A poor outfielder such as Jack Cust becomes viewed as a DH, relegating him, in all likelihood, to DHing in Triple-A.

Some of those reputations are deserved; others aren't. However, the point is only that bad defenders in the minors are easy to overvalue: if, say, Elvis Andrus's future is in left field, his future isn't as bright as it is if he proves he can make the grade at shortstop.

It isn't yet clear how minor league defensive numbers translate to their equivalents at the big-league level, but it seems like a safe bet that the overall quality of defense goes up. Thus, a player who is below average in the minor leagues isn't likely to contribute with the glove when he earns a promotion to the show - that is, if he stays at that position at all.

Yesterday, we looked at the best minor league fielding performances of 2006. Today, let's turn to those guys whose gloves may keep them from advancing, or whose performance may send them spilling down the wrong side of the defensive spectrum. As I pointed out yesterday, these numbers aren't a final judgment on each player's defensive skill - they are simply a good estimation of how well he performed this past season. Young players can still improve, and lady luck can reverse itself.

The lists are generated by the same method as the "Best of" lists in yesterday's article. IP is the number of innings played at the position (there's a minimum of 600 for inclusion), PAA is the number of plays above an average fielder at that position in that league, and PAA/150 is plays above average per 150 games. Of course, for the worst fielders, PAA and PAA/150 will be negative. Once again, the numbers aren't adjusted for park, league, or level.

Second Base

Player            Level   Org      IP      PAA    PAA/150
Edgar V Gonzalez  AA/AAA  Flo      697.3   -36     -70
Hector Pellot     A       Nym      803.3   -38     -64
Micah Furtado     A+      Tex      629.3   -29     -63
Felix Molina      AA      Min      852     -31     -49
Corey Wimberly    A+      Col      682.3   -23     -47
Isaac Omura       A       Oak      632     -20     -44
Nate Spears       A+      Chc      711     -23     -44
Dan Dement        AA      Was     1064.7   -34     -43
William Bergolla  AAA     Cin      820.3   -25     -42
Jeff Natale       A/A+    Bos      741     -23     -42

Few of the names at the top of this list have much of a future with their bats, so these fielding numbers may be the final nails in each coffin. Bergolla's defensive skills will probably keep him from having much value in a utility role, and Natale's - as has been predicted since he was drafted - will surely lead to a move away from second base. The notable prospects who are significantly below average are Kevin Melillo of the A's, Martin Prado of the Braves, and Hernan Iribarren of the Brewers.

Third Base

Player            Level   Org      IP      PAA    PAA/150
Koby Clemens      A       Hou      705     -28     -54
Chase Headley     A+      Sdp     1099     -32     -40
Bryan Bass        A+/AA   Bal     1000.3   -28     -38
Brian Snyder      A+/AA   Oak      846.3   -23     -37
Ryan Braun        A+/AA   Mil     1008.7   -27     -36
Ryan Barthelemy   A+      St.      638.7   -16     -34
M. Vechionacci    A/A+    Nyy     1112.3   -28     -34
Matthew Brown     AA      Laa     1123.3   -27     -33
Josh Fields       AAA     Chw      973.3   -24     -33
Mike Kinkade      AAA     Flo      615     -13     -31

Again, these numbers match up with conventional scouting analyses in a couple of big-name cases. Braun is a good bet for an eventual move to the outfield; Fields has already begun working on a possible transition. Braun's stats over the course of the year make a case for either the volatility of defensive stats or the difference between Single-A and Double-A. He was slightly above average in his half season in the Florida State League, but in fewer than 60 games at Double-A, he was 28 plays below average. Other prospects on the wrong side of average are Eric Campbell (-27 PAA/150) and Matt Tuiasosopo (-21).

Shortstop

Player            Level   Org      IP      PAA    PAA/150
Diory Hernandez   Rk/A+   Atl      664     -33     -67
Sergio Santos     AAA     Tor     1116     -53     -65
Matt Smith        A       Tex     1039     -47     -62
Ian Desmond       A+      Was      768.3   -31     -56
Eduardo Nunez     A/A+    Nyy      831.3   -34     -55
Alcides Escobar   A+      Mil      689     -27     -55
Chris Nelson      A       Col      951.7   -35     -50
Jeffrey Dominguez A/A+    Sea      868.3   -31     -48
Chris McConnell   Rk/A    Kan      967.3   -31     -43
Matt Maniscalco   AA      Tam      887.3   -27     -41
Elvis Andrus      A       Atl      894     -26     -39

When the Blue Jays traded Orlando Hudson and Miguel Batista for Troy Glaus and Sergio Santos, they couldn't have made a much more lopsided deal, at least so far as defensive performance is concerned. In addition to Alcides Escobar, there are a number of fielders whose gloves have gotten raves but did not score well with Range: Erick Aybar and Chin-Lung Hu are a bit below average, Asdrubal Cabrera came out at -36 PAA/150, and Ben Zobrist ended up at -30.

Center Field

Player            Level   Org      IP      PAA    PAA/150
Jerry Gil         AA/AAA  Ari      800     -36     -61
Kevin Mahar       AA      Tex      609     -22     -49
Juan Senreiso     A+      Kan      825.3   -25     -41
Clay Timpner      AA/AAA  Sfg     1079.7   -31     -39
Joe Holden        SS/A    Nym      841.7   -24     -38
Trevor Crowe      A+/AA   Cle      722.7   -20     -37
Austin Jackson    A       Nyy     1089     -28     -35
Jeff Salazar      AAA     Col      712.3   -18     -35
David Murphy      AA/AAA  Bos      891.3   -23     -35
Austin Jackson    A       Nyy     1097     -28     -34

Matt Kemp and Reggie Willits both scored just below average in this metric, but other than those two, there aren't a lot of discrepancies between conventional scouting wisdom and this year's center field numbers. Fernando Martinez made 7 fewer plays than average in just over 600 innings, but his future is probably in a corner. The only surprise on this list is that a center fielder could conceivably be bad enough to cost his team more than two wins with the glove - and keep getting starts in center.

Left Field/Right Field

Player            Level   Org      IP      PAA    PAA/150
Mitch Jones       AAA     Nyy      735     -30     -55
Shin-Soo Choo     AAA     Sea      681.7   -27     -53
Brian Mc Fall     A+      Kan      920.7   -36     -53
Daniel Carte      A       Col      954.7   -34     -48
Ryan Harvey       A+      Chc      998.7   -34     -46
Delwyn Young      AAA     Lad     1088.3   -37     -46
Brian Pettway     A       Tor      872.7   -29     -45
Xavier Paul       A+      Lad      993.3   -33     -45
Garrett Guzman    A+/AA   Min      905.7   -30     -45
Sergio Pedroza    A       Lad      760.3   -25     -44

Some of these players may well bounce back to be average, or slightly below-average defenders, but it's worth noting that these ten outfielders - plus Sergio Pedroza, Chris Lubanski, and a few others - were all worse with the glove than Jack Cust. More shocking than that is that Lubanski saw 225 innings in center field. I use the verb "saw" because it appears that's all he did: in that time, he was 17 plays below average, resulting in -104 PAA/150, a performance so bad I don't have an adjective for it. Other familiar names within spitting distance of the Cust line: Jeffrey Corsaletti, Ryan Patterson (who also, inexplicably, saw 141 innings in center, with Lubanski-like results), Nolan Reimold, and Billy Butler.

By Way of Conclusion...

If you're interested to see more, these numbers are now available for every 2006 minor league player at MinorLeagueSplits.com. In another year or two, with additional full seasons of play-by-play data, it will be easier to make confident claims about the defensive skill of minor leaguers. It also may be possible to analyze the effects of position switches, so when all of these players become corner outfielders, we can predict whether they'll give Chris Lubanski - or even Jack Cust - a run for his money.

Comments

Good stuff Jeff. These are basically range factors, right?

I just bookmarked you guys.

Austin Jackson appears twice for center field. Thought I'd point that out.

And interesting that Vechionacci, whom scouts seem to like his D from what I've heard, shows up as bad defensively here.

You really can't make any true assumptions with these ratings. Several of those players are very good defensively. You also have to remember that field conditions in minor league facilities, especially the lower levels, are usually not very good and can vary greatly from one to another. Add in the lights issue (some fields have very poor lighting), and you have to almost ignore minor league defensive numbers and go by the scout's opinions. Rarely are scouts wrong defensively, because it's the on area where they can see players perform on a daily basis and it's not dependent on the players around them.

I understand and applaud the attempt, but I would start making assumptions on the defensive ability of players based off these numbers.

I understand and applaud the attempt, but I would start making assumptions on the defensive ability of players based off these numbers.

I think that's why Jeff's conclusion mentions how he can't wait to have more seasons worth of data to work with in order to better assess minor league defensive performance.

Rarely are scouts wrong defensively, because it's the on area where they can see players perform on a daily basis and it's not dependent on the players around them.

That's completely ridiculous. The difference between the best and worst defenders is about one play every FIVE games. Scouts rarely get to watch players play on a "daily basis", at least not most players, or for very long. Scouts MAY be better at noticing things than average fans are, but the difference is so elusive that scouts can and will be wrong.

And...not dependent on players around them? Equally ridiculous. If you're playing third base next to a brilliant shortstop, there are going to be plenty of plays the shortstop will make that you should've. If you're playing next to Mr. Brilliant, nobody notices that you should've made the play--it's a 6-3. If you're playing next to...let's see...Sergio Santos, it's a single to left, and those trustworthy scouts are saying, "you should've made that play."

These are basically range factors, right?

No, it's quite a bit more advanced than that. It uses play-by-play data and predicts the number of batted balls that should be fielded by each defender. Here's an introduction to the stat:

http://www.hardballtimes.com/main/article/measuring-range/

Could a 27 or 28-year old in AA or a 26-year old in A ball who don't have major league talent score higher than a younger player with more potential just because of their experience?

Jeff Sackmann is taking on one of the final frontiers of analyzing baseball stats, and I look forward to his long-term results.

Sackman,
You need to get away from your computer and join the real world my friend. Your stats only approach is what gives stat geeks a bad name. Also, your work here is absolutely ridiculous. Stats are a tool to be used, they are not the end all like you suggest. I'll take a scouts opinion on defense over any defensive rating based off minor league statistical numbers. This analysis is flawed. Stick with your website. At least that has merit. You are in over your head here.

Re the comment directly above, Jeff never claimed these stats or rankings were the "end all." In fact, he wrote, "these numbers aren't a final judgment on each player's defensive skill - they are simply a good estimation of how well he performed this past season."

I believe this is an excellent start, and I hope Jeff will continue to run the numbers to shed light on what heretofore has been an area with little or no coverage.

I don't know if this explains the 2nd Austin Jackson, but I noticed this past weekend that Colby Rasmus came out very poorly on this system of tracking plays, as well. He could be just off the bottom, since his numbers are -28/-34 in 1104 Inn. I was surprised he rated so poorly, but am guessing that augers a move to a side field.