College flychasers

There are a lot of obstacles in the path of quantifying the defense of college outfielders. Defensive analysis is challenging enough as it is, and compounding the difficulty is the short length of the NCAA season and the limited availability of college stats of any kind.

I can’t make the NCAA season any longer, but having built a Division 1 play-by-play database going back to the 2007 season, I can use as much of that data as possible to take a step toward quantifying the defensive skills of those who have played college ball in the last three seasons. Even if you have no interest in the amateur game, you may find it instructive to compare the college performance of some recent draft picks to their results in the pros.

The process

The method by which I generate defense numbers is similar to that of many other play-by-play defensive metrics, such as Sean Smith’s TotalZone. The main differences are related to the more limited data available for college. For instance, while I know whether outs are groundballs or flyballs, I don’t know whether any specific hit resulted from a ball on the ground or in the air.

So we make use of what we have. Given all of the plays for which an outfielder was playing his position, I calculate the mix of lefty-lefty, lefty-righty, righty-righty, and righty-lefty matchups, as well as the ratios of groundball outs to flyball outs for the pitchers on the mound. This data across all of Division 1 provides the average number of plays a fielder at a certain position makes. Counting the number of plays a fielder made and then comparing that to the average results in a number: the “plus-minus” common to many systems.

One important note: My college numbers are denominated in plays, not runs (as is TotalZone, for instance). If you want to think of these numbers in terms of runs, a rough translation would be to multiply the number of plays by 0.8. If the outfielder is playing shallow and picking off bloop singles, the multiplier might be less. If he plays deep and regularly spears home run balls over the fence, it might be more.

If all of this sounds familiar, it may be because it is. I used a similar process last year to calculate defensive numbers for college third basemen and shortstops.

Reliability

Given the short season and the limitations of the data, it’s to be expected that the data contains lots of noise. While we can come up with a reasonable estimate of how many fly balls were “catchable” by an outfielder given the handedness of certain batters and pitchers, along with the flyball tendencies of the pitchers, it’s just that: an estimate.

With three years of data, we have lots of numbers on outfielders who played back-to-back years, against different hitters, often behind different pitching staffs, and sometimes alongside different fellow fielders. Using these pairs of year-to-year numbers for players who were in the field for 40 or more games each season, it appears that about 25 percent of their raw plus-minus numbers are based on talent. The other three-quarters can be ascribed to luck.

While that much luck isn’t a huge vote of confidence in any statistic, it’s about as good as we could hope for given the constraints of measuring college outfield performance. (In fact, I anticipated worse.) To keep the numbers in a believable range, reflecting as much skill and as little luck as possible, every college number you see below is regressed approximately 75 percent. This amount varies a bit depending on playing time; an outfielder who barely crossed the 40-game plateau is regressed more; one who went all the way in the College World Series and played up to 60 games is regressed less.

Outfielders to watch

With the Division 1 season getting underway this weekend, I hope many of you will take the opportunity to catch a game, especially in these several weeks where college teams are playing baseball that counts long before the pros do. With that in mind, here are some of the best active collegiates, according to their results from 2008 and 2009.

Let’s start by looking at center fielders:

Player           School            +/-  
Mummey Trent     Auburn            +10  
McGuiggan Steve  Illinois Chicago   +9  
Moore Jonathan   Purdue             +9  
Malloy John      Lasalle            +8  
Parker Jarrett   Virginia           +8  
Holt Tyler       Florida State      +8  
Martin Jason     San Jose State     +8  
Heid Drew        Gonzaga            +7  
Rowe Connor      Texas              +6  
Johnson Addison  Clemson            +6

These are the leaders among players who were in center for at least 800 balls in play (about 40 games) in both 2008 and 2009. To put each fielder on the same scale, the plus/minus numbers shown are per 2000 balls in play (about 100 games).

It’s gratifying to see someone like Auburn’s Trent Mummey at the top of the list. Mummey won a 2009 ABCA/Rawlings NCAA Division I Gold Glove Award and was named to last year’s SEC All-Defensive Team. One of the great things about fielding stats is that they can unearth quality defensive performers who don’t get rewarded with accolades like Gold Gloves. But before we can gain enough confidence in the sleepers an algorithm digs up, it’s nice to have confirmation that, in a broad sense, the statistic quantifies something similar to what scouts are looking for.

It’s also interesting to note the number of top overall prospects like Tyler Holt, Jarrett Parker, and Connor Rowe. This list is likely to include bigger-name players for a couple of reasons. First, the playing time minimum means that active players must have been full-time players in their freshman or sophomore seasons. Second, the lower level of specialization in college baseball means that, in a very general sense, a baseball player who is good at one ballplaying skill is more likely to be good at others.

Of course, quality fielding isn’t limited to center field. Again looking at active players for the 2010 college season, here are the best corner outfielders over the last two seasons:

Player         School              +/-  
Keen Reggie    Radford              +8  
Roberts Aaron  Southern Illinois    +7  
Lewis Chris    Western Michigan     +7  
Heere Brian    Kansas               +6  
Lee John       South Dakota State   +6

It may not be a surprise that there’s less a bit less distinction at the top end among corner outfielders. Given the lower overall level of talent in college, very good outfielders tend to play center. Further, there are fewer plays to be made in the corners, especially in cases where there’s an athletic center fielder making all the plays next door.

Draftees and their pro numbers

As I’ve noted, many of the best defensive outfielders are also top-notch hitting prospects. Just as Tyler Holt and Jarrett Parker are likely to go in the first few rounds in this year’s draft, so some of last year’s top outfielders have already logged time in the pros.

This gives us an opportunity to compare college and minor league defensive numbers. Given the sample sizes involved (100 or so college games; a partial season in the low minors), this isn’t nearly enough to calculate any sort of translations between college and pro fielding, and some of these numbers should be treated as little more than curiosities. But it is a way to further confirm that a certain player does or doesn’t excel in the field.

Here are some of the top defensive center fielders from 2008-09. Only this time, the list is limited to those who were drafted and spent some time in the pros in 2009. Minor league numbers are TotalZone ratings, denominated in runs saved.

Note that Dustin Ackley isn’t included as he hasn’t made his pro debut. For the curious, his outfield results were almost exactly average.

Player           School             Team     +/- (NCAA)  +/- (MiLB)    Mi Pos  
Rockett Michael  Texas San Antonio  Tigers          +14          +6     LF/RF  
Smith Rand       Appalachian State  Marlins          +5          +1        CF  
Jackson Brett    California         Cubs             +5          -1        CF  
Nommensen Brett  Eastern Illinois   Rays             +5          -5     CF/RF  
Harrilchak Cory  Elon               Braves           +4          +5  LF/CF/RF  
Kipnis Jason     Arizona State      Indians          +4          +5     CF/LF  
Freitas Nick     Southern Utah      Twins            +4          +4        CF

Given the amount of regression that constitutes part of these numbers, it’s utterly amazing how many plays Michael Rockett made to be a +14 center fielder at UTSA. In limited time in the Tigers organization last year, he put up numbers nearly as strong, though at corner outfield positions. With the exception of Brett Nommensen‘s weak pro debut in the field, this sampling suggests that defensive success in Division 1 translates to the pros, and that none of the stats involved are too wonky.

Here are some top fielding corner outfielders who came out of last year’s draft:

Player                School           Team       +/- (NCAA)  +/- (MiLB)    Mi Pos  
Cavazos-Galvez Brian  New Mexico       Dodgers            +5          -7  LF/CF/RF  
Mitchell Jared        Louisiana State  White Sox          +5          -5        CF  
Henry Jordan          Mississippi      Indians            +5          +9        CF  
Liddle Steven         Vanderbilt       Twins              +4          +1     LF/RF

In another limited sampling, the relationship between college and minor league numbers is all over the board. Most striking is the difference between Jared Mitchell and Jordan Henry. Mitchell went from a strong college defensive career in a corner to a weak pro performance in center. Henry had similar success in college but looks fantastic in his first pro shot at center field.

The 2008 draft

If we go back one more year, we have even more data to look at. I have defensive ratings back to 2007, so for a 2008 college draftee, we can look at two years of data and, in most cases, two years of minor league data as well. Here are the 10 top picks from the ’08 draft for which I have sufficient data:

Player            School        Team      D1 CF  D1 LF/RF  Mi CF  Mi LF/RF  
Hudson Kyle       Illinois      Orioles      +4       n/a     -2         0  
Ratliff Sean      Stanford      Mets          0       n/a      0        -2  
Russell Kyle      Texas         Dodgers     n/a        +8     -5       +26  
Kieschnick Roger  Texas Tech    Giants      n/a        +4    n/a       +12  
Blackmon Charlie  Georgia Tech  Rockies     n/a         0    +10        +1  
Raben Dennis      Miami (Fl)    Mariners    n/a         0    n/a        -3  
Carroll Sawyer    Kentucky      Padres      n/a        -1    n/a       +18  
Schafer Logan     Cal Poly      Brewers       0       n/a     +7        +1  
Cowgill Collin    Kentucky      D'Backs      -1         0     +5        +1  
Tekotte Blake     Miami (Fl)    Padres       -3       n/a    +13        -1

As with the last table, these numbers are all over the board. Limited comparisons such as these are mostly for fun; even if we looked at every player who crossed over from college to the pros in the last three years, we wouldn’t have enough data to calculate translations with very much confidence.

However, one thing stands out. Notice all those plus signs in the minor league columns! Surely the +26 and +18 from Kyle Russell and Sawyer Carroll are flukes, but in all probability that indicates that something well above average is going on. Roger Kieschnick, Charlie Blackmon, and Blake Tekotte show double-digit positives, as well.

There is a possible explanation. In their first pro season or two, college outfielders are playing in the low minors alongside high school, junior college, and Latin American products. Whether the 22-year-old outfielder has the most potential as a major leaguer is beside the point. By the time they reach the Single-A level in the minors, they have received more years of coaching, and their polish reflects that. To get some idea, consider that Kieschnick is eight months older than Justin Upton.

More flychasing

As is usual with amateur baseball research, one project invites several more. To complete our picture of the defensive skill of college outfielders, it would be valuable to look at their success holding and gunning down baserunners. It may be valuable to break Division 1 into parts to get a better idea of “average” in elite and non-elite conferences. And of course, when more data is available down the road, it will be instructive to calculate the relationship between college and minor league results.

In the meantime, I’ll be making the extra effort to see Trent Mummey and Michael Rockett in action. See you there!

Print Friendly
 Share on Facebook0Tweet about this on Twitter0Share on Google+0Share on Reddit0Email this to someone
« Previous: The compilation: MLB team-by-team top-10 prospects
Next: Waiver Wire Offseason: NL »

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>