How much stock should the Astros put in Tommy Manzella's glove?

This piece has to start with a strong preface: I don't know the answer to this question, nor will I by the end of this.  All I can use is the information available, spotty, at best, due to the inherent shortcomings of defensive metrics and their required sample size.  I can also rely on scouting reports and my own first hand account watching Manzella play defense in Round Rock.  These, however, are all ways at getting the truth, but never actually arriving.  It'll be an approximation, and that leaves plenty of room for error.

That said, I want to try and get to a general idea of how much probability we should be assigning to Tommy Manzella's glove.  It's something we've discussed numerous times in the comments section this offseason, but it's something that hasn't receive the full attention of a front page article.  It's something that's been brewing about in the back of mind for sometime now, but it's not something I've had the heart to actually sit down and quantify—or at least attempt to do so. That's the rub of being a blogger about your favorite team, you have to look at the ugly truth about your team far too often.

Now, I probably have a personal bias against Manzella's fielding because an un-scooped mis-throw of his almost took my head off back in April when I was in a camera well at Dell Diamond.  But aside from the fear that either myself, or my friend's $600 camera, was going to get clobbered because of the errant throw, the rest of what I've seen of Manzella hasn't inspired a lot of confidence.  My honest impression of seeing him play is that he has the range, but his arm isn't accurate enough; again, though, there's a huge potential bias on my end in terms of his throwing prowess.

I doubt you're reading this to read through my internal monologue about whether I'm being too harsh on Manzella's defense.  So, we'll get to the numbers.  We'll primarily be taking a look at TotalZone, a metric that was conceived of, and honed, by Sean Smith (the CHONE guy).  Though not as advanced as UZR, its simplistic input allows for prevalent minor league data, which is what we need for this endeavor.

Obviously, as Astros fans, our logical benchmark for a great defensive shortstop is Adam Everett.  Tommy Manzella is not the next Adam Everett just because I'm not entirely convinced we didn't watch the best fielding short stop ever in 2006.  Failing to live up to the impressive glove work of Everett doesn't mean Manzella won't be a good, or even great, fielder.  Just that I want to see if we can come to a more concrete way of predicting his success than what the Astros organization has touted about Mr. Manzella.  So, after the jump, I'll give you my best effort at coming up with just that.

(If you're interested in what Manzella's bat might look like, relative to Everett's, click here for AstroAndy's excellent investigation of the subject last summer)

There are several ways of viewing Tommy Manzella's defense that have kept me up the last few nights thinking them through. The first, and somewhat valid, method would be look what Manzella's marginal impact our defense will be from 2009 to 2010.  The process is pretty straight forward: Take Miguel Tejada's TotalZone Rating from 2009 and then using what we know about Manzella's defense determine what the marginal change would be.

Miguel Tejada, in 2009, had a TotalZone rating of -22, which we can approximate as runs.  Not a stellar year for Miggy, and a drop of twenty-three runs from 2008.  Since 2005, Tommy Manzella has posted yearly TotalZones of 2, 0, 2, 3, and 3 (via MinorLeagueSplits.com).  Over a five year period, he averages a TotalZone of 2.  I'm comfortable with assuming that Manzella's five year average is a reasonable—enough—approximation for his defensive skill. Thus, the marginal expected change in the Astros defense, at shortstop, in 2010, is 24 runs.  That's an impressive change and would likely offset whatever Manzella's batting line is compared Tejada's.

However, it doesn't really answer the question we need answered.  We don't want to know if Manzella will be better than Tejada at defense.  What we want to know is if Manzella is plus a defender.  Plus enough to offset a dismal/triple/slash, that is.

Our mythical benchmark is Adam Everett. And, even as unlikely that I feel it is that we'll see another Adam Everett-esque defensive shortstop again (until Jiovanni Mier, of course...), let's run with that concept.

After scouring the internet far and wide for what I could find in terms of defensive metrics for minor leaguers, I'm depressed.  The goal was to find something that I could use to compare the development of Everett and Manzella using apples to apples.  Unfortunately, the best statistic that's out there for evaluating minor league defense, Total Zone, can only go back so far (2005, to be exact).

I did, in the search, learn more about the utility of TotalZone ratings for minor league players.  And it is something we should address before we push onward:

For the ratings to be useful, they need to correlate from one level to another. If we knew a player had a +15 rating one year, that would be of no use unless it told us that he was likely to continue to post good ratings in future seasons, at higher levels of the minors. Preliminary investigation shows that these ratings are more useful for infielders than outfielders. The correlation is much lower for outfielders, though at least they are (usually) positive. I’ve found that you can usually get a correlation of 0.50, meaning you regress 50 percent to the mean, at about 350 chances for infielders. This represents less than a full season of chances for second, third and short. For outfielders, you need about two full seasons of data to regress 50 percent, or about 1,000 chances.

The only season's worth of data that we have about Manzella where he has 350+ chances at a single level is 2009; the same season in which he posted his best career number of +3.  That's something to consider as the other years that reported rely on incomplete samples as Manzella changed levels.

The other thing to consider is that Sean Smith has developed an MLE (major league equivalent) for TotalZone based on his regression analysis of defense from every level.  For a SS, the MLE is -10 at AAA.  However, these MLE's are based on 500 chances, aren't park adjusted, and are based flyball/linedrive tendencies in the minors.  If we scale the MLE to Manzella's 400 chances at AAA, though, he'd be rerated as a -5.  Still a full 18 runs better than Tejada was last year, but certainly no Adam Everett.

This has been a lovely digression, and something that should be variable in our evaluation of Manzella's defense, but let's get back to the task at hand.  The only apple to apple comparison I could find for Manzella and Everett was Range Factor (click here to read Tango's critique of the metric).  As imperfect as it is, it's what we have to work with. And, truth be told, fielding metrics are imperfect, so we could also just choose to view Range Factor as just being more imperfect and sweat it a little less (although, it looks like things are getting pretty damned close to accurate: here and here).

Manzella looked as though he were in a career arc of improvement, until 2009 saw him post a step backwards.  I'm wiling to give him the benefit of the doubt in 2009 because John Gall was his first baseman (he's no Puma, we'll just leave it at that). There could have very well been plays that Manzella would have gotten credit for if he had a better first baseman, but the fact remains, Manzella remains a step behind Everett no matter how we slice it. 

If we were to draw any conclusions on Manzella's defense, I'd say the following:

• At every level, Manzella is behind Everett.
• Manzella, however, does seem to be gradually improving, much the same way that Everett did—just never quite reaching Everett's height.
• The prior to observations leads to a defensible conclusion that Manzella will always be a step behind (below?) Everett.
• That conclusion doesn't necessarily impose a low ceiling for Manzella, though.

First, there's the fact that Manzella has a been a perennial C prospect for John Sickels, who writes that Manzella is "a steady and reliable defensive shortstop.  His bat isn't that bad, so that leads me to infer that really isn't bringing a truly plus glove, otherwise he'd be more than a C.  BPro's, Kevin Goldstein had this to say of Manzella, "[a] fundamentally sound shortstop." I don't read that and get goosebumps.  Of course, Goldstein follows that with "[h]e's a fantastic glove man whose instincts and 55 speed (on the 20-80 scale) give him plenty of range. Manzella also has outstanding hands and an above-average arm." Mixed signals from Goldstein, no love from Sickels...not the greatest start.