February 27, 2007
Gold Gloves in February
As Nate mentioned in Unfiltered, Rawlings is conducting fan balloting for an all-time Gold Glove team. Last fall, after doing an ESPNews segment on the topic with Hall of Famer Ozzie Smith, I was contacted by Rawlings and asked to be part of the panel that determined the ballot for this project. I was honored to do so.
Clay Davenport sent me reams of data on the best fielders in the Gold Glove era. Clay's defensive ratings, which account for context better than any other, are the best non-play-by-play metrics extant, and they formed the basis for most of my balloting. Where I needed more information, such as when I had to fill out more ballot spots than were covered by Clay's data, I tended towards selecting players of recent vintage, whose excellence I can support with performance data, observational evidence and a greater understanding of their reputation.
With the balloting out, I figured I'd poke my head in and lay down a vote. I honestly didn't expect my nominees ballot to be reflected well on the final ballot. I found, in doing this, that there was a wide gap at many spots between defensive reputation-best measured by Gold Glove awards-and defensive performance as measured by Clay's data. I'm glad that performance analysis had a seat at the table, even if Jesse Barfield or Glenn Hubbard or Barry Bonds, all of whom have terrific defensive numbers, didn't quite make the cut. It is interesting to note that performance and reputation do match up in a lot of spots. Keith Hernandez, Bill Mazeroski, Ozzie Smith…the greatest defensive players of all time show up atop the leaderboards in Clay's system.
I'm getting ahead of myself. Let's go through the ballot. Once again, I will be using Clay's metrics primarily in voting:
Oddly, I don't have aggregate data for pitchers' defense, and I initially went with Kaat, Maddux and, I believe, Mike Mussina. Of the three Gibson shows up as slightly below average (-12 FRAA), Kaat slightly above (9 FRAA), and Maddux blows them both away (77 FRAA). Kaat has been considered the best-fielding pitcher ever, but it's pretty clear that the honor now belongs to Greg Maddux. He gets my vote.
As Rob Neyer pointed out, while it's nice that they've gone ahead and listed a lot of guys at some of these positions, most will come down to one or two players garnering the most support. This is one of the more interesting cases, as Bench has been surpassed by Rodriguez over the past two seasons, lapping Bench in FRAA (199 to 181) as he's also passed him in games caught.
Bench will have his supporters, given his importance to the defining baseball team of his era. Beyond that, I think you can make an argument that Bench's throwing prowess was more important to the Reds than Rodriguez's has been to his teams, because of the greater emphasis on the stolen base in the 1970s. While Clay's numbers account for run environment, they don't take into account "stolen base environment."
Having to choose, I would choose Rodriguez, in no small part because he's still an above-average catcher (15 FRAA in 2006) adding to his lead on Bench. However, the difference in the responsibilities of a catcher in the '70s and the '90s is enough to at least make you think about leaving Bench on top, despite what the numbers say.
For a teenager in New York in the 1980s, Hernandez vs. Mattingly was our version of Willie, Mickey and the Duke. It turns out that it shouldn't have been much of a debate; Mattingly's Gold Gloves and reputation notwithstanding, he doesn't show up well in performance metrics. All defensive measures of first basemen miss one particular skill-handling throws-that is critical to the position. While Mattingly was very good at this, and had to be with those 1980s Yankee infielders, Hernandez was just as good and arguably better.
Hernandez's 173 FRAA bury everyone else on the ballot. In fact, the five spots after him are occupied by players who don't appear on the ballot: John Olerud, Mark Grace, Jeff Bagwell, Pete O'Brien and Norm Cash.
There may be no position where reputation and performance diverge the way they do at second base. Ryne Sandberg was an above-average fielder by most metrics, worth 87 FRAA in Clay's system, squeaking into the top ten. Roberto Alomar? Ten FRAA in his career, barely above-average. Debates about Alomar's defense were raging when I first discovered rec.sport.baseball in 1991, and they continued throughout his career.
The top of the list is easier to handle. Bill Mazeroski is one of two players in the Hall of Fame strictly for his defense, and it was great defense. His career mark of 258 FRAA (253 in the "Gold Glove era") is head and shoulders above the pool. Frank White is second with 155.
What's interesting here is that Ken's brother Clete Boyer, probably one of the three greatest defensive third basemen ever, isn't listed. Playing as a contemporary of Robinson, he won just one Gold Glove award, that in 1969 after he landed with the Braves.
Robby is the greatest gloveman at third by acclamation, courtesy his amazing showing in the 1970 postseason. He's also the greatest statistically, about 30 runs better than Schmidt over the course of their careers. Scott Rolen, at 149 FRAA, has an outside chance to catch Robinson, although he's slowed a bit the past two years.
Ozzie Smith isn't just the greatest defensive shortstop ever--he's the greatest defensive player in baseball history. His mark of 271 FRAA is the most of any player, with only Mazeroski all that close. It's adorable that they listed five other guys on the ballot with him, especially Jeter, who's got a LaToya Jackson-in-"We Are the World" thing going on here.
You can at least fake an argument at second base for Frank White, and the problems with measuring first-base defense statistically create cases for some players at first base. At shortstop, though, any vote that isn't for Ozzie Smith is just silly.
Outfield: Paul Blair, Roberto Clemente, Andre Dawson, Jim Edmonds, Dwight Evans, Curt Flood, Ken Griffey Jr., Torii Hunter, Andruw Jones, Al Kaline, Garry Maddox, Willie Mays, Kirby Puckett, Ichiro Suzuki, Larry Walker, Devon White, Dave Winfield, Carl Yastrzemski
That is one odd group of candidates, a reflection of the truism that the easiest way to win your fifth Gold Glove award is to win your first. (Don't get mad, guys, just look at the results.) Defensive performance fluctuates much, much more widely than you'd know just from looking at the awards, which are based on reputation more than performance.
The highest-scoring outfielder in Clay's system is Jesse Barfield, and I'm not smart enough to understand why. He had a short career and a cannon arm, and from 1985-1988 was worth 89 runs with that arm and good range, and 134 in his career.
If you fill out the ballot the way Gold Glove voting is typically done, with all outfielders lumped together, you get Flood, Mays and Blair, three amazing defensive center fielders. Andruw Jones would be next. If you elect to choose three outfielders by position, you get Flood or Mays (Flood leads 136 to 124, and technically, Mays accumulated a lot of that value in the years before Gold Gloves were awarded), Barfield and Barry Bonds. Well, you don't, because neither of the latter two are on the ballot. Barfield I can understand, but Bonds won eight Gold Gloves, and every outfielder with at least six except him and Dwayne Murphy (six exactly) is on this ballot, so I'm going to go out on a limb and say that that's not quite right.
With the two best candidates not on the ballot, I'd vote for the three center fielders. There's not a left fielder on this ballot worth voting for ahead of them.
I think this is a fun fan exercise, and I was happy to have a small part in it. While I expect the final results to reflect reputation and popularity, it is interesting to note that at the very top of the pyramid, the actual performance of players is pretty much in line with those elements.