Table of Contents

Tuesday, June 05, 2007

May 2007 Reds Review - Part 2: Defense

As we saw in part one, the Reds' defense continues to be one of the largest problems facing the team. Tonight, I want to investigate that in more detail. Let's start with a figure comparing two measures of team defense: Dave Studeman's Runs Saved statistic (+-runs) and Michael Lichtman's UZR fielding stats, summed across players on each team. Stats current through May 31st for this season for NL Teams:
Comparison of THT's Runs Saved statistic and total UZR Runs Saved for teams. The latter was calculated by summing UZR scores for all Reds. Both statistics report runs saved above average, and assess the combined effect of range and sure-handedness. More than 70 runs separate the best and worst teams, which accounts, by itself, for 7 wins in the first ~55 games. The diagonal line indicates a perfect match between THT and UZR.

The story is not pretty. The Reds' defense is clearly among the three worst teams in the National League. And given that UZR is based on more detailed data than the THT data, there's a legitimate argument that the Reds have the worst defense in the league. The average of these two estimates puts the Reds at -27 runs below average. Already.

A quick thought exercise: If the Reds "just" had an average defense this season, that would mean that they would have given up 259 runs this season (instead of 286), or 4.7 runs per game. This would rank them 10th in the league in runs allowed, up from their current ranking of dead last, and would improve their PythagoPat record to 27-28. And with a top-flight defense, saving 40 runs above average like the Cubs (UZR) or Mets (THT)? That's 77 fewer runs, giving 3.8 runs allowed per game, which would be second only to San Diego (PythagoPat improves to 32-23). Defense matters.

I'm surprised by how bad the Reds defense has been. Shocked might be a better word. I expected the Reds to be better this year on defense than in 2006. Gonzalez taking over SS from Lopez and Clayton, Griffey's move to right field, and Ryan Freel (who was stellar in CF last season) and/or Josh Hamilton in center field all should have improved the defense, while the other players remained constant.

Instead, it appears to have actually gotten worse, at least so far this season. According to THT's +- runs stat, the Reds' defense was -20 runs below average all of last year combined. Here, it is only in early June, and they have already surpassed that mark. What on earth is going on??

Thanks to MGL's extremely kind decision to release his UZR statistics through May 31st of this season, we can look more closely at team defense. I don't have full details on how UZR is calculated, but you can think of it as normal zone rating, just adjusted for a variety of important effects such as park factors and (I think) handedness of batters. Since THT isn't yet able to post its ZR stats, I don't have a good fielding stat to compare UZR to right now (the ZR reported by ESPN isn't a very good version of ZR), so I'm just going to report the data as a table:
Name POS
Team Runs Saved (Range) Runs Saved (Fielding Errors) Runs Saved (Other Errors) Total Runs Saved G
Conine, Jeff 3 cin 0 0 0 0 15
Hatteberg, Scott 3 cin 1 -1 0 0 16

Phillips, Brandon 4 cin 2 1 0 3 38
Freel, Ryan 4 cin 0 0 0 0 1

Keppinger, Jeff 5 cin 1 0 0 1 3
Castro, Juan 5 cin -1 0 0 -1 3
Freel, Ryan 5 cin 0 -1 0 -1 11
Encarnacion, Edwin 5 cin -7 -2 0 -9 21

Castro, Juan 6 cin 0 0 0 0 4
Gonzalez, Alex 6 cin -2 1 0 -1 37

Hopper, Norris 7 cin 1 0 0 1 5
Hamilton, Josh 7 cin 0 0 0 0 3
Dunn, Adam 7 cin -4 -2 0 -6 31

Hopper, Norris 8 cin 0 0 0 0 5
Freel, Ryan 8 cin -6 0 0 -6 27
Hamilton, Josh 8 cin -6 0 0 -6 20

Hamilton, Josh 9 cin -1 0 0 -1 12
Hopper, Norris 9 cin -1 0 0 -1 1
Griffey, Ken 9 cin -3 -3 0 -6 33

Let's go position-by-position. Unless I mention otherwise, all fielding stats I mention are UZR, though you can look at my 2006 fielding review for more info on last year's fielding performances.

1B: Both Conine and Hatteberg have been average at first. Hatteberg, at least, was better than average last season, so we might hope for improvement there.

2B: Phillips is above average this season at +3 runs, which is a very nice improvement over last season. UZR has Phillips as below-average last year, though most other range stats pegged him as right about average.

3B: Ugh. Here's about a third of the Reds' struggles. For all the fuss about Eddie's tendency to make errors, those only are knocking his UZR rating down by two runs. The rest is bad range. Horrible range, in fact. Eddie is currently last among all third basemen in both his total rating of -9 runs and his range rating of -7 runs. He's on pace to smash his -20 runs rating of last season.

I really do think he's better than that--he did post a very positive UZR his first season with the Reds (2005). And he's young, so he can improve. But we can't wait forever. Fortunately, my impressions are that he's been better since his return from AAA--Marty certainly was pleased with him during the last home stand, and he's about as down on Encarnacion as anyone can be. I hope he can get himself to be at least an average 3B by the time 2008 rolls around.

SS: Gonzalez is currently rated a thoroughly average -1 runs saved. Interestingly, despite already making more errors this season than last, he's actually above average in error rate for his position and below average in range (both only slightly). Hopefully he can get back to the form that allowed him to save +9 runs above average last season, as he is a good defender, even if he's overrated by the media--he is clearly not anywhere close to being in Adam Everett's league. Nevertheless, he remains a significant defensive upgrade over Felipe Lopez (-17 runs last season).

Oh, a fun aside: you know how the media kept talking about how Boston was going to miss Gonzalez's glove this season? His replacement, the supposed lead-gloved Julio Lugo, has thus far saved +4 runs above average, five more than Gonzalez. ... 'course, he's hitting 0.226/0.289/0.330, so that part of the equation isn't working out well yet.

LF: Dunn, at -6 runs below average, is on pace to break his personal worst of -20 runs below average that he set last season. Dunn is far from the worst left fielder in baseball (Manny Ramirez is -13 runs below average already), but he doesn't help us much out there. He probably should be playing first base, but we have Joey Votto coming up at that position. So barring a trade, we'll just have to accept him and deal with it. I don't expect much improvement, at least not any better than his 2006 performance.

CF: WTH? The big surprise to me is Ryan Freel, who is rated at -6 runs below average. Last season, other metrics (THT's ZR and PMR) rated him well above average defensively in center, so I'm having a hard time knowing what to make of this. Freel was rated as +5 runs saved per 150 games last season (+1 run saved in 45 games) by UZR, so this is a major drop-off, even if UZR doesn't "like" him as much as other stats. Hopefully he can come back from his head injury, rediscover some speed or something, and start running down in fly balls.

Hamilton, as I discussed in my piece on him for THT, is probably playing out of position. His range is decent, but best suited to right field, where his arm can continue to help win the occasional game. I do support trying to trade Griffey sooner rather than later (more on that in another post), and that would open RF for Hamilton. But the Reds will likely continue to be shaky at center field for the foreseeable future.

RF: Griffey hasn't had a very good start to his time in RF. Encouragingly, though, his range is only -3 runs below average, which is certainly tolerable for someone as productive as he has been at the plate. The rest of his -6 runs has to do with his errors, which were notable earlier in the season but seem to have tapered off. I think he'll be fine.

In sum, the only players in the Reds' defensive lineup (catchers excluded) who seems to be performing at or above the level we expect of them are Brandon Phillips and (maybe) Adam Dunn. That's the bad news. The good news is that I expect all of those other players to be better over what's left of the season. Nevertheless, continuing to improve the Reds' defense over the long term has to be one of the biggest priorities for Wayne Krivsky as he looks toward this years' trading deadline and the Reds' return to greatness in 2008. :)

I'm going to give them short-shrift, at least for now (sorry, past bedtime). But the primary way research has shown that catchers can consistently contribute to a teams' defense is via their ability to stop the running game. This season, Dave Ross continues to be a standout, throwing out 46% of all would-be base stealers. Moeller, on the other hand, continues to show a weak arm, throwing out just 14%. The surprise, however, is Valentin, who has yet to throw out a base-stealer in 11 attempts. Last year, he caught 8 of 18, and he has a career 32% caught stealing percentage, so this is very unusual for him. I almost wonder if he might be having arm problems. If so, it's the first good reason that I've seen for the Reds to carry three catchers...if only Moeller could throw out runners, he might actually be a good fit.