In Defense of the Braves?

Much ink has been spilled this winter over the various and sundry changes the Braves have had to make in their starting lineup and with their pitching staff; these changes have been substantial and have deserved every drop of that ink. But what about another, less talked-about aspect of the Braves' game?

I'm talking about defense. The front four of the rotation doesn't strike anybody out and doesn't give up many homeruns; depending on who the Braves ticket as the 5th starter, the same thing might well be said of the entire starting rotation. With a staff full of pitchers who are bound to put more balls in play than most previous Braves' staffs, team defense becomes all the more crucial.

But how do we measure defense? We could use our eyes, but that's an obviously flawed "method"; we don't see nearly enough of the play on television to get a truly solid idea of how it developed. Fielding percentage is decent enough if you want to figure out how prone to errors a given player is, but when it comes to measuring the full defensive ability, it's really pretty weak.

You're probably already aware of Range Factor and Zone Rating, two more advanced defensive metrics cooked up by Bill James and STATS Inc, respectively. We'll let Baseball Primer's gifted sabrmetrician Mitchel Lichtman define and discuss these two statistics:

"Bill James introduced us to Range Factor (RF) as, essentially, the number of outs made per game. This was convenient at the time, because we had no other context but the game. The problem is that a game is not necessarily nine innings for each fielder. As well, each fielder is dependent on his pitching staff and "luck" for opportunities.

STATS began tracking Zone Rating (ZR) as, essentially, the total number of outs per balls in a fielder's "area of responsibility" (i.e., zone). This addressed some of the shortcomings of RF. However, STATS ZR has many shortcomings of its own. For example, each fielder is only given one zone. We know that it's much easier to convert a ball in play into an out if the ball is hit near you, rather than on the fringes of the zone. "

(Note: Lichtman's analysis can be found here,)

So what can the interested Braves' fan use if he wants to look at his beloved team's defense? Once again we turn to Mitchel Lichtman.

In the above-linked article, Lichtman introduces his "new" defensive system, which he calls "Ultimate Zone Rating", or UZR for short. A more complex description and discussion of the system is available in the link for the more intrepid fan. But basically, UZR factors in dozens of variables which heretofore have not been included in more sophisticated defensive metrics, including park factors, flyball/groundball tendencies of pitching staffs, even the speed of a batted ball. It is an exciting contribution to the landscape of defensive evaluation. Let's use it to look at the Braves defense.

(Note: UZR does not rate catchers, so that switch will not be included in the analysis. The numbers I quote in this column can be found here. Note also that those numbers are three-year averages.)

Second Base

Marcus Giles:One run above average per 162 games over the last three years

The knock on Giles coming up through the minors was always his glove; it was assumed that he would seriously struggle defensively at the big league level.

That concern statistically appears to be overwrought, as a lot of folks suspected. On the other hand, what doesn't seem exaggerated is the level of Giles' defensive improvement in 2003. He looked like he had gone from below-average to sheer brilliance; UZR tells us he was two runs above average per 162 games last year, compared to 18 runs below average per 162 in 2001. The 2002 statistics look shockingly good, but come from a very limited sample size. It should be noted that, like the 02 numbers, his 2001 UZR extrapolations are derived from fairly limited time.

He did make an extraordinarily impressive change though, and as long as he stays motivated to prove wrong the hundreds of scouts and observers who questioned his abilities, there's no reason to assume he'll regress.

Judgment: Solid defender


Rafael Furcal: 4 runs above average per 162 games over the last three seasons.

OK, so he's frustatingly inconsistent during the season, and we all hold our breath when he's getting in position to field an easy double play grounder. He was at seven runs above average per 162 games in 2003, and his other years look relatively consistent. Furcal's a darn fine defensive shortstop; it's easy to forget that at times with his schizophrenic fielding tendencies.

Judgement: Good defender

Left field

Chipper Jones: One run below average per 162 games over the last three seasons

Bill James has written hundreds of incredibly clever, incisive things over the span of a great career as a commentator. But possibly the most intelligent observation he's made is that teams have an unfortunate tendency to blame their ills on their best players. Like Cronus or the French Revolution, organizations eat their best products out of an unfortunate illusion about their player's weaknesses. Teams focus too much on what the player can't do and not enough on what he can.

In fairness to the Braves, they've been pretty good at avoiding this when it comes to Chipper Jones. They moved him to left field in 2002 for Vinny Castilla, but aside from that they've treated their star well. They gave him a contract extension, never publicly attacked him during his occasional slumps and always seemed to have a good idea that his contributions with the bat far outweighed the detriment that was his glove.

But unfortunately, James' principle can also be applied to fans, respected journalists and Jay Mariotti. Chipper's defense at third was never really bad enough to warrant the derision it received; he made errors at bad times during the 2000/2001 seasons, and that condemned him in many fans' eyes. Like those who obsessively concentrated on Greg Maddux's 199 1/3 innings in 2002 or his refusal to pitch to Javy Loepez instead of looking in awe at a Hall of Fame career, too many Braves' fans have honed in on Chipper's supposedly awful defense and have begun to tear down a player who should be considered a franchise great.

Look at the numbers above. In his left field career, Chipper Jones is one run below average. One run. You'd think that would be a laughably small price to pay for a career .309/.404/.541 hitter, but evidently not.

Yes, Chipper was worse in 2003, 12 runs below average per 162 games played. That's very bad. But as Lichtman would almost certainly point out, year-to-year fluctuations don't necessarily mean the things we think they mean. It could just be statistical noise, or regression to the mean. (Chipper was very good in 2002) The larger sample size tells us that Chipper Jones is basically an average left fielder with an excellent bat. Maybe we should stop focusing on what happened on the last day of the 2000 season and start looking at the bigger picture.

Judgement: Average defender

Center field

Andruw Jones: 15 runs above average per 162 games over the last three years

A lot of what was written in the Chipper Jones' comment can be cross-applied to Andruw Jones. We have this image in our head of Andruw doing truly miraculous things while making them look routine; doing a Spiderman routine on the Turner Field wall to rob Jay Bell, effortlessly gliding to balls no other center fielder could catch. And now we have this image in our head of an overweight, downright slow Jones, chugging after balls he routinely caught not long ago. The former image is accurate but is now several years old; the latter image isn't at all true, but both hurt Andruw's reputation among Braves' fans.

Again, the numbers tell the story. Andruw Jones is a thoroughly excellent defender. He was brilliant in 2003 after a disappointing 2002, but again, remember that those fluctuations don't necessarily mean much. What we have in Jones is one of the game's best defensive center fielders. That really should be enough.

Judgement: Excellent fielder.

Right Field

2004: J.D. Drew: 8 runs above average per 162 games over the last three years.

2003: Gary Sheffield: Six runs below average in right field per 162 games over the last three years.

You'll notice that the format for this position is a little different. It's the only position so far listed where the Braves will undergo a change in players. (First base and third base will be discussed, together, at the end of the column)

Sheffield never looked particularly comfortable or athletic in right field, and the numbers seem to bear that out. He wasn't awful out there, and his bat more than made up for his glove, but the defensive upgrade here is fairly substantial.

Two things to keep in mind about Drew's numbers: First, they're extrapolated from a small sample size due to Drew's injury problems the last several seasons. Secondly, those same injury problems are almost certainly depressing Drew's stats. A healthy J.D. Drew, while heretofore a contradiction in terms, would almost certainly post better numbers than that. Judgement: Substantial defensive upgrade First base, third base, and a conclusion Robert Fick wasn't most Braves' fans favorite player by the end of the 2003 season. He finished ice cold with the bat, and his glove seemingly betrayed the Braves down the stretch and in the playoffs. Indeed, he did hurt the Braves twice in the LDS against the Cubs with his iron glovework.

But the bigger picture is much more kind to Fick; he's at one run above average per 162 games over the last three years. For the Braves, that stat was exactly average. So how much of an upgrade defensively can we expect from a Julio Franco/Adam LaRoche platoon?

Franco's good reputation and pleasing to the eye defensive play seem legit; after a rough 2001 with the glove, Julio has sparkled, playing at 20 runs above average per 162 games played in 2002 and at 13 runs above average in the same stat during the 2003 season. Sample size issues apply here of course; Bobby Cox has been pretty careful about spotting Franco against lefties, and a weight-lifting injury kept Julio on the DL for a while last year. But still, those numbers give cause for hope. According to UZR, the best first basemen over the last few years have been Todd Helton and Doug Mientkiewicz, at 20 and 17 runs above average per 162 games played respectively.

Adam LaRoche is a bit harder to evaluate since he hasn't played an inning at the major league level. Scouting reports on his defense are universally exuberant; no one is down on his ability to pick it an first base. Scouting reports can be flawed of course, but the near unanimity in the praise for LaRoche's defense lends credence to the idea that he's a fine defensive player. Let's be conservative and slot him in at the Scott Spiezio/Derrek Lee level of defensive play: five or six runs above average per 162 games played.

Fick wasn't really that bad in 2003, but the combination of Franco and LaRoche promises to be a decent upgrade defensively.

Third base is a little more challenging to evaluate. Vinny Castilla certainly has never been a stathead favorite, and with good reason. His defense was supposedly overrated by analysts who focused too much on errors and not enough on range.

Well, Vinny's now at 4 runs above average per 162 games played over the last three years, which obviously is pretty good. Castilla was at about just that level in 2003, another rebound from a sub-par 2002.

Where this gets a little difficult is trying to evaluate Castilla's replacement. Mark DeRosa appears to be the man right now, but Wilson Betemit, Russ Branyan and Michael Hessman all have shots to take over third base should DeRosa falter. DeRosa doesn't look to be a particularly good third baseman; Mark doesn't have enough games at third base over the last three years to make his way into the link provided, but Lichtman's spread-sheet extrapolates DeRosa's 2003 defensive performance to eight runs below average per 162 games played. Small sample size warnings obviously apply here.

On the other hand, Branyan's 2003 third base performance projects out to 31 runs above average per 162 games played, so that's pretty cool. It's also patently ridiculous, a measure of small sample size issues. In 2002 he was at 69 runs below average per 162 in Cleveland and 40 runs below per 162 in Cincinnati. His 2001 numbers were similarly…brutal, too say the least. I'm not smart enough to figure out what all of that noise means for Branyan. I'm inclined to look at his 2003 numbers as some sort of bizarre fluke; on the other hand, his 2002/2001 numbers can't possibly be accurate. They would make him the worst defensive player in baseball, and I can't believe that. Still, it's hard to find a way to call Branyan anything but a bad defensive third baseman.

All in all, the Braves look to have a solid club defensively. First base, second base, shortstop, center field and right field all look to be solidly above average in 2003. The only position where the Braves are looking at a defensive sinkhole in third base, and that's a little hard to quantify.

Atlanta won't be brilliant defensively, not enough to carry the team if one of the bigger elements suffers. But the Braves can look forward to a season where their defense won't present much of an encumbrance to their hopes of success.

Atlanta Dugout Top Stories