Designated Hitter November 30, 2006
The Best Minor League Defenders

Prospect analysis has always been one of the most contentious issues in the "stats vs. scouts" debates. However, for all the breakthroughs in statistical techniques, analysts have almost always had to rely on scouting to assign a value to the defensive contribution of a young player.

No longer. Through its website, Minor League Baseball has made available a play-by-play log of every game played in the affiliated minors, complete with some batted-ball information. A person with enough time, desire, and misdirected energy can track every ball that was pitched, hit, or caught by a bush leaguer in 2006.

For pitching and hitting, there's MinorLeagueSplits.com. We can turn our attention, then, to fielding. Using a statistic called Range, I came up with plus/minus ratings for every 2B, 3B, SS, and OF in the minors. (For more information on how Range is calculated, here's David Gassko's introduction to his creation.)

I haven't yet calculated the run values of plays at various positions for each league, but you can estimate them if you wish. A single is worth about half a run, so if you figure every extra play that an infielder makes saves a single, divide the number of plays by 2 and you've got runs above average. (Odds are the infielder--particularly if he's a third basemen--saved a few doubles too, so actual runs saved is somewhat higher.) Outfielders save more runs per play; a higher percentage of the plays they make would otherwise turn into doubles and triples, which are worth about 0.72 and 1.04 runs, respectively.

Before looking at some of the best fielding performances in the minors, I feel obligated to spend a couple of paragraphs hedging. Note I used the word "performance." There's plenty of luck in just about every baseball stat, and that goes double for defensive stats. A good fielding performance means that a defender made more plays than you'd expect him to. It doesn't mean he's the next Adam Everett (though he could be!), or that he'll even be above average next year, just that he had a great year.

Another lacking--and this is something I'm working on--is that all of these stats are relative to league average. It seems like a reasonable assumption that an "average" shortstop in Triple-A is much better than his "average" counterpart in Single-A. Along the same lines, an average fielder in Triple-A would be below average in the majors. Many players improve in their first few years in organized ball, and more importantly, those who aren't good enough either get cut or move to a less demanding position.

Finally, these numbers aren't park-adjusted. With only one complete year of data to go on, defensive park factors would be almost as unreliable as the numbers they would adjust. Keep that in mind when you see some of the staggeringly high play totals for corner outfielders.

What follows is a top-ten list for the minor leagues at every position I've done so far. Each list shows the number of innings played at the position (IP), the number of plays made above average (PAA), and the number of plays above average per 150 games (PAA/150). The cutoff for inclusion is 600 innings played at that position; in most cases, the stats you see reflect the player's entire season, regardless of whether they were promoted or traded at some point. With all of that out of the way, let's see some numbers!

Second Base

Player            Level   Org      IP      PAA    PAA/150
Mario Holmann     A/A+    Nyy      727.7    49      91
Brian Cleveland   A+/AA   Flo      674      33      66
Fernando Cortez   AAA     Kan/Tam  699.3    29      56
Joshua Johnson    A       Kan      865.7    35      55
Eric Patterson    AA/AAA  Chc     1147      41      48
Drew Sutton       A+      Hou     1016.7    36      48
Luis Cruz         AA      Sdp      701      24      46
Jayson Nix        AAA     Col      867.7    29      44
Brooks Badeaux    AA/AAA  Bal      649.3    21      44
James Guerrero    A       Flo      608.3    19      41

After Eric Patterson and maybe Jayson Nix, this is not a list of attention-getting names. Indeed, there aren't a whole lot of big-name 2B prospects in the minors; many future second basemen still consider themselves shortstops. That said, Dodger farmhands Blake Dewitt and Tony Abreu both look solid: each are in the top 20 of all minor leaguers, with 20 and 22 PAA/150, respectively.

Third Base

Player            Level   Org      IP      PAA    PAA/150
Jonathan Malo     SS/A+   Nym      707.7    37      71
Edward Lucas      A+      Kan     1036.7    46      59
Patrick Cottrell  A+      Tam      898      37      55
Kevin Kouzmanoff  AA/AAA  Cle      633.7    24      51
Brennan King      AAA     Phi      821.7    30      49
Vince Rooi        A+      Pit      929.7    34      49
Ian Stewart       AA      Col     1006.3    34      46
Nick Petrucci     A       Cle      752.3    24      44
Pat Osborn        AA      Cle      614.3    20      43

Even after trading Kevin Kouzmanoff, the Indians are stacked. Five more spots down the list is Andy Marte at 36 PAA/150. However, in the neighborhood are also some names that may lead you to question how valuable an above-average AAA defender is: Earl Snyder, Mike Hessman, and Fernando Tatis are among the top 30, above 20 PAA/150. Either Snyder and Hessman have gotten bad raps that relegate them to the Quad-A all-star team, or the difference between MLB-quality and AAA-quality hot corner defense is huge.

Shortstop

Player            Level   Org      IP      PAA    PAA/150
Ramiro Pena       A+/AA   Nyy      673.3    36      72
Kevin Hooper      AAA     Det      669.7    35      71
Agustin Septimo   A       Flo      631      33      70
Jonathan Diaz     SS      Tor      625      32      70
Agustin Septimo   A       Flo      644.3    32      67
Leonardo Acosta   Rk/A    Chw      693.7    32      62
Brent Lillibridge A       Pit      624.3    28      60
Brian Bixler      A+      Pit      623.7    27      57
Andy Cannizaro    AAA     Nyy      879.3    36      55
Oswaldo Navarro   AA      Sea      694.3    26      50
Jesus Gonzalez    A       Tor      796.7    28      48

Meet your future utilitymen. If any of these guys can consistently play shortstop at this level, there ought to be a roster spot for them someday. (Though Andy Cannizaro might be best off finding a new organization.) Of the big-time SS prospects in the minors, Reid Brignac, Brandon Wood, and Dustin Pedroia had the best defensive seasons in 2006, each in the neighborhood of 10 PAA/150. On the basis of their performance this year, Troy Tulowitzki, Erick Aybar, Chin-lung Hu, Stephen Drew, and (surprise) B.J. Upton will need to improve substantially to play a decent big-league shortstop.

Center Field

Player            Level   Org      IP      PAA    PAA/150
Jacoby Ellsbury   A+/AA   Bos      914      44      65
Justin Upton      A       Ari      918.3    32      47
Brent Johnson     A+      Sea      847.3    28      45
Dustin Majewski   A+/AA   Tor      856      28      44
Tony Gwynn Jr.    AAA     Mil      812.7    26      44
Antoan Richardson A       Sfg     1000.7    29      39
Chris Amador      A+/AA   Chw      805.7    23      39
Sam Fuld          A+      Chc      734      21      38
Yordany Ramirez   A+      Sdp      638.7    17      37
Matt Young        A+      Atl      747      20      36

I can't wait to work out MLB equivalents for these; if Ellsbury's true talent level is even half of his 2006 PAA/150, he could be one of the best defensive centerfielders in baseball right now. It's tough to get quite so excited about Justin Upton just yet, but Diamondbacks fans who would like to see him stick in center have to be encouraged by his ranking here. Just out of the top 10 is another major prospect, Carlos Gomez of the Mets, who put together 32 PAA/150.

Other prospects who seem to deserve their reputations are Chris Young (28 PAA/150), Michael Bourn (26), Cameron Maybin (21), Felix Pie (19), Adam Jones (18), and Fernando Perez (16).

Left Field/Right Field

Player            Level   Org      IP      PAA    PAA/150
Tim Battle        A/A+    Nyy      899.3    56      84
Drew Anderson     AA/AAA  Mil      953.7    55      78
Quentin Davis     A       Atl      736.3    42      77
Charlton Jimerson AAA     Hou      685.7    34      67
Jonel Pacheco     A       Nym      889.3    33      50
Travis Buck       A+/AA   Oak      749.7    27      49
Leyson Septimo    A       Ari     1111.3    38      46
Brian Gordon      AAA     Hou      672      22      44
Cody Strait       A+      Cin     1144.7    37      44
Doug Clark        AAA     Oak      750.3    23      41

Since so many minor leaguers bounce around the outfield, especially when they switch teams, there are lot more qualifying corner outfielders if I put the innings totals together. As I mentioned above, some of these numbers are ridiculously high for corner guys--do you really think Tim Battle was worth 3 or 4 wins with the glove? It's always possible: in 233 innings in center, Battle racked up another 12 plays above average.

Of the familiar names who had good defensive seasons, Adam Lind sticks out; there seem to be plenty of questions about his glovework, but he put up 28 PAA/150 in 2006. Also respectable were Nelson Cruz (38, good for a spot in the top 20), Luke Scott (22), and Hunter Pence (18).

By Way of Conclusion...

There's a lot of work left to be done to know just how much these numbers mean, but if nothing else, it's nice to have some statistical corroboration for some common scouting reports. It will probably take another season or two of data to come up with accurate park adjustments and comparisons between each level and the major leagues. For now, we can already spot some potentially undervalued players, and raise our eyebrows at the raves granted to others. Standing at the edge of a new frontier is always exciting, even if you can see nothing but clouds.

Jeff Sackmann is a weekly columnist for The Hardball Times and the creator of MinorLeagueSplits.com, the only source for comprehensive split and situational stats for every active Minor Leaguer. He is a long-time Brewers fan and blogs about them at BrewCrewBall.com. He also contributes to the fantasy magazine Heater and the football site PackerBackerBlog.com. Jeff lives in New York City, where he earns his keep helping people get into business school.

The surprise is that so much of this opposes what we hear. So who can tell... I'm somewhat skeptical it's a clear cut case either way. We know Eric Patterson has been described in unflattering terms defensively, Kouz as a great defender is probably news to 99% of people. I've actually seen Brent Lillibridge play, and while numbers mean more than my own definitely-not-a-scout eyes, from what I saw and what I read he didn't look like a shortstop wizard. Also surprising is that Asdrubal Cabrera, called by many "the best defensive shortstop in the minors" is nowhere to be found.

Sam Fuld as a better defensive center fielder than all the guys not listed ahead of him seems downright shocking.

I have to say I'm a little surprised by the results given the scouting reports I've read. In addition to the APing's comments, I was under the impression that Chin-lung Hu was a great fieling, no-hit SS.

In addition to previous commenters, this sentence made me question the stat's validity: "On the basis of their performance this year, Troy Tulowitzki, Erick Aybar, Chin-lung Hu, Stephen Drew, and (surprise) B.J. Upton will need to improve substantially to play a decent big-league shortstop." Tulo, Aybar and Hu -- even Drew, really -- are all praised as being very good defenders. Better than those listed above them, without question. Guessing, but since this statistic only uses play-by-play data, we aren't really seeing range incorporated. I just gotta question it.

And really, Adam Lind, Fuld, and Kouzmanoff are good defenders?

Sure looks like Ellsbury is special in the field. I just wonder if his bat will be enough to get him to CF in Boston to show off.

Thanks for covering a topic that has long been overlooked. Because of all the variables, defensive ratings for minor leaguers is an area that still has much to be discovered.

All the Tulowitzki scouting reports I've read peg him has merely adequate at SS.

I watched Tulowitzki with the Rockies and he has decent range, but there was one thing that bothered me about him. He is the opposite of smooth when throwing the ball to first base. I don't know if he's trying to create velocity because there's a lack of it, but every throw to first on a DP is capable of going into the stands.

I watched Tulowitzki play dozens of games in college, and I would grade his range as above average and his arm as a plus (even at the MLB level). I would be surprised if he didn't end up being at least an average defensive SS in the bigs.

Stephen Drew was fourth from the bottom in David Pinto's Probabilistic Model of Range this year:

2006 Shortstop PMR

Oh, and I've heard good things about Kouzmanoff's defense from various places, so it didn't come entirely out of nowhere. Nice work Jeff.

I'm curious as to where Alex Gordon ranked. By all subjective anecdotal accounts, he is a fantastic defender. Do the stats bear that out?

He's a ways down the list, but well above average.

http://www.minorleaguesplits.com/pl/460/460086dfs06.html

Stephen Drew was fourth from the bottom in David Pinto's Probabilistic Model of Range this year

You mean, the same PMR that ranked Bill Hall second among all MLB SS? If the man who hit 35 HR last year was truly the second-best defensive SS in baseball, do you really believe he would be #3 on the depth chart and looking more like the starting LF next year?

My point isn't to demean PMR or any of the other fielding metrics. David Pinto's work is valuable. Jeff Sackmann's work is another breakthrough. And on and on. Thanks to these systems, we know more about evaluating defensive performance today - and with greater confidence - than ever before.

However, in the spirit of recognizing that no one system is fool proof, we need to continue to question the inputs and outputs rather than taking at face any of these findings. We can - and will - do even better if we remain open minded.

We can - and will - do even better if we remain open minded.

I commented solely for the purpose of adding to the open minded atmosphere in the comments section, since someone had questioned Drew's ranking so low.

Hall is not the starting shortstop because his bat can be useful elsewhere on the field, and the Brewers are not done seeing what J.J. Hardy can do with the glove and with the bat as a shortstop. Also, I can see the Brewers dealing Hall for some extra help, considering they already have players waiting in line for the corner outfield, 3B and SS positions. Having him play in the outfield for a bit is a decent strategy for raising his trade value a bit, since there are plenty of teams that need a 3B, SS or corner outfielder, especially one making so little money in comparison to the rest of the bloated contracts handed out this offseason.

So yeah, it's possible.

Also, Hall's moving to the outfield is a front office decision, and may not necessarily reflect the true ability level of the players in question. Really now, do the very best hitters always play where and when they are supposed to? Of course not, and that's much easier to evaluate than defense is.

Hall is not the starting shortstop because his bat can be useful elsewhere on the field

That doesn't even make sense. Bats don't play in the field. If Hall is a better defensive shortstop than Hardy, then he should be the SS. Shortstop is a much more important position than left field. Put Hall at SS and Hardy in LF (or someone else). The fact that Hall is a better hitter than Hardy shouldn't have any bearing on where you would place them in the field.

The only reason you would play Hardy over Hall at SS is because the former is a better fielder than the latter, not because he is a worse hitter.

Furthermore, putting Hall in LF does not increase his trade value as much as leaving him at SS, particularly if he is as good as the PMR would suggest. Making him a LF would only serve to reduce his trade value.

Hall is not the second-best defensive SS in baseball. He's probably not even the second-best SS on the current MIL roster. The Probabilistic Model of Range got it wrong. If the PMR can get it wrong on Hall, then it can also get it wrong on Drew and others.

As I said before, my point isn't to demean PMR or any other system. As a tool, they are valuable. But they are not the definitive word either - at least not at this point.