AaronGleeman.com

Guzman's 2004 season was one of the most confusing I've ever witnessed as a baseball fan, because he was really not significantly better than he was in past years, when he was criticized quite often, but the difference this year was that he got tons of praise from both Twins fans and the local media.

Here's what Guzman has done offensively over the last three years:

YEAR       G      AVG      OBP      SLG      OPS      GPA     VORP

2002     148     .273     .292     .385     .677     .228     11.7

2003     143     .268     .311     .365     .676     .231     12.7

2004     145     .274     .309     .384     .693     .235     14.8

Not only wasn't Guzman's 2004 season significantly better than the two years that came before it, the consistency of his mediocrity is remarkable. What the Twins have gotten from Guzman over the last three years is a guy who will miss 15-20 games a year with various injuries and benchings, hit an incredibly empty .270, draw one walk a week, and produce 10-15 runs more than a replacement-level shortstop.

His defense is where it gets interesting. I personally thought that Guzman's defense was improved this year, but that he was still average, at best. Several defensive metrics agree that he was improved, but instead have the improvement being huge.

YEAR     FRAA     FWS

2002       -1     7.7

2003       -5     7.0

2004       17     9.6

"FRAA" stands for Fielding Runs Above Average, a Baseball Prospectus stat that attempts to measure how many runs a player was worth defensively over an average player at the same position. This is a stat in which Guzman made a massive improvement, going from -5 and -1 in 2002 and 2003 to +17 this year.

"FWS" stands for Fielding Win Shares, which is a Bill James creation (as part of his overall Win Shares) that we track over at The Hardball Times. Like with FRAA, Guzman made a major improvement in Fielding Win Shares, going from 7.7 and 7.0 in 2002 and 2003 to 9.6 this year, which actually led all major-league shortstops.

The problem, as has been explained to me by people with much bigger brains, is that stats like FRAA and Fielding Win Shares simply estimate the number of opportunities a player gets defensively, which limits the accuracy of the numbers. There are, however, other defensive metrics that examine actual play-by-play data for exact figures, rather than estimates.

The most mainstream of those is Zone Rating, which measures the percentage of balls hit into a player's defensive "zone" that the player makes a play on. Guzman's Zone Rating this year (.823), while up up, ranked just ninth among everyday AL shortstops and was not meaningfully higher than his Zone Rating last season (.821), when no one praised him for anything other than the quality of his wizard-beard.

Over at Baseball Think Factory, Chris Dial has a defensive metric called the Dial Player Index that uses the Zone Rating numbers that are generated from play-by-play data. DPI has Guzman as improved this season, going from -15 and -11 in 2002 and 2003 to -7 this year, but still well below average. As Chris told me last night when I asked him for his figures for Guzman: "He did play 'better,' but still poorly."

Another defensive metric, Mitchel Lichtman's Ultimate Zone Rating (UZR), uses detailed play-by-play and hit-location data for real opportunity numbers, but then takes it a step further by attempting to adjust for the difficulty of each opportunity. Guzman was awful in 2002 and 2003, checking in at -19 and -17 per 150 games, but is at +2 in Lichtman's initial numbers this season (he told me he still has some adjustments to make before releasing his "final" 2004 numbers).

This is a perfect example of the problems many people, myself included, tend to have with defensive stats. Even the best of them -- and the ones I've mentioned are definitely very good -- sometimes come up with numbers that can't help but bring up major issues. In this case, my issue is with the fact that, prior to this season, Guzman's defense was shown to be very poor by nearly every advanced defensive metric.

Now the numbers show that he made a complete turnaround in just one year, going from horrible to great in FRAA, average to great in Fielding Win Shares, and horrendous to above average in UZR. Only DPI shows him as still being below average. Now, it's certainly possible that those numbers are legit, but I have a hard time believing it's just a coincidence that Guzman's gigantic improvements came in the same season that the old, fast Metrodome turf was replaced with new, slower turf, and the Twins went from having a fly ball pitching staff to a ground ball pitching staff.

Perhaps in past years Guzman was unfairly penalized for having few opportunities (because of the fly ball pitching staff) and those opportunities being very difficult ones (because of the fast turf). Or perhaps now Guzman is being unfairly credited with having lots of opportunities (because of the ground ball staff) and those opportunities being easy ones (because of the slow turf). Or maybe it's neither of those things, but either way something just doesn't seem right to me.

How does a guy go from being -22, -20, -19 and -17 in UZR from 2000-2003 to being +2 in 2004? How does he go from being -12, -15, -8, -1 and -5 in FRAA from 1999-2003 to being +17 in 2004? How about from being in the middle of the pack in Fielding Win Shares to leading all of baseball? I suppose I might be willing to write it all off to legitimate improvements or a "career year," if not for the changes in turf and the Twins' pitching staff.

All of which is a very long way of saying that it is no longer easy to say that "Guzman stinks offensively" and also add in "and defensively too." It might still be true, but there is now some doubt. Hell, it might never have been true, if you really want to give Guzman the benefit of that doubt. When all else fails, of course, I go back to what my admittedly untrained eyes tell me, which is that Guzman is, at best, an average defender.

There is no doubt that Guzman has the skills to be a very good player, and he has certainly shown that potential many times during his six years with the Twins. However, he turns 27 next year, he doesn't seem like a particularly hard worker, and he doesn't strike me as someone in any danger of making huge strides as a player. At this point, he is what he is: a slight above/below average shortstop, both offensively and defensively, who does enough things well to make you want to kill him when he makes a mental error, bounces a lazy lob throw over to first base, or pops up a bunt in a key spot.

The Twins have a shortstop prospect named Jason Bartlett, who hit .331/.415/.472 at Triple-A this year and .296/.380/.425 at Double-A in 2003. I'm probably a bigger Bartlett fan than most people and I think he will eventually be a very solid everyday middle infielder in the majors. However, I'm not certain that his future position is shortstop and I'm even less certain that he would be able to step in for Guzman next year and perform adequately.

What the Twins need to weigh is just how far away they think Bartlett is and, if he's not ready, just how much of a dropoff Nick Punto and Augie Ojeda would be from Guzman at shortstop. Yet another consideration is who they might be able to get cheaply on the free agent market. Is Guzman worth $2 million more than those options? How about $3 million? They've clearly decided that he's not worth $5.25 million more, which, as I told people all season despite plenty of disagreement, was a no-brainer.

I'd love to see the Twins sign Omar Vizquel to a one-year deal, I wouldn't mind seeing Guzman back for $2 million or so, and I'd have no problem going into next spring with Bartlett, Punto, Ojeda and perhaps a cheap veteran free agent battling for the everyday shortstop job, assuming the money saved could be spent wisely somewhere else. Ultimately, what Terry Ryan does between now and the start of next season will determine how people view the decision to cut Guzman loose, but choosing not to devote 10% of the team's payroll to him is smart move regardless of what happens next.