The defensive All-Stars

The All-Star ballot I outlined in an earlier post took only batting into account.

I didn’t project a better-rounded team because there aren’t established metrics that everyone can agree on for fielding or baserunning.

Fielding has Range Factor, a fairly simple metric that tells part of the story. By merely dividing total chances by games (or 9 innings), you can get a measure of how much ground a player covers. That way you can differentiate a player who can get to a lot of batted balls from another who makes plays on balls hit close to him.

What it can’t tell you is whether a player’s Range Factor is higher because he has more balls hit in his direction than other players at his position. For example, if he’s a third baseman on a team with three or four left-handed starting pitchers and a large number of ground-ball pitchers.

Other people will tell you about DWAR, Defensive Wins Above Replacement, even after your eyes glaze over from boredom and an aversion to math.

First basemen and catchers don’t have measurable range. For 1Bs, I’m using DWAR. Catchers do have a viable measure, catcher’s ERA, in addition to DWAR.

By combining the available metrics, I came up with this list of defensive All-Stars that I voted for on a ballot in addition to my "true" offensive All-Stars:

American League

C Salvador Perez, Royals

1B Mike Napoli, Red Sox

2B Dustin Pedroia, Red Sox

3B Josh Donaldson, Athletics

SS Alexei Ramirez, White Sox

LF Yoenis Cespedes, Athletics

CF Leonys Martin, Rangers

RF Nick Markakis, Orioles

National League

C Yadier Molina, Cardinals

1B Matt Adams, Cardinals

2B Brandon Phillips, Reds

3B Pablo Sandoval, Giants

SS Troy Tulowitzki, Rockies

LF Khris Davis, Brewers

CF Marcell Ozuna, Marlins

RF Jason Heyward, Braves

Three players are on both my offensive and defensive All-Star teams. I don't think you could come up with a good reason not to vote for Donaldson, Ramirez and Tulowitzki. OK, you're going to tell me Donaldson has more errors than anyone west of Pedro Alvarez, but at the time I checked the defensive metrics, Donaldson led major league third basemen in DWAR and the AL third sackers in Range Factor.

Rockies 3B Nolan Arenado and Athletics RF Josh Reddick would have been on my defensive ballot, but they're injured.

There were some ties. 

With only one metric, DWAR, at first base -- and precious little positive value, with just four 1B above 0.0 when I checked the numbers, there was little choice. Napoli tied with Kansas City's Eric Hosmer and Adams with San Diego's Yonder Alonso. I broke the tie the way Gold Glove voters seem to, by which is the better offensive player. In hindsight, I should have gone with Hosmer in the AL. He is a truly gifted defender. Don't talk to me about the Diamondbacks' Paul Goldschmidt. I haven't seen him in person very much, but when I have he has struck me as a good hitter but not a good fielder. At the time I looked at DWARs, Goldschmidt was at -0.5, or a greater negative value than any first baseman's positive value.

Speaking of defense, I just noticed that I made an error at second base in the NL. The Rockies' D.J. LeMahieu ranked ahead of Phillips in both RF and DWAR.

For the NL catcher, Molina and the Giants' Buster Posey had the same combined ranking. As a tiebreaker, I used DWAR, where Molina ranked higher, ahead of CERA, where Posey led the league.

Similarly, Ozuna's superior DWAR broke a tie with the Cardinals' Peter Bourjos and Heyward's major league-best DWAR in right field put him ahead of Marlins RF Giancarlo Stanton. By the way, lost among some pretty good offensive statistics, Miami has probably the majors' best outfield defense. LF Christian Yelich also ranked second in the NL in the combined metrics.

* * *

I also looked for individual base-running metrics that could help in establishing a list of well rounded All-Stars, but there didn't seem to be enough evidence (call it small sample size) to make an educated guess about one player over another. Baseball Prospectus' base-running metric (positive and negative runs) includes five different types of runs -- Ground Advancement, Stolen Base, Air Advancement, Hit Advancement and Other Advancement.

There could be some merit in the larger sampling of team base-running runs. At the time I researched this post, the Royals led the majors with almost 10 runs, 50% more than the runner-up Mariners. The Cubs led the NL. By far, the worst teams were the Red Sox and Athletics, with the old and infirm Phillies last in the National League. 

KC's base-running prowess could be a reason why the Royals stayed in contention despite some atrocious hitting. Base-running appeared to be just another problem for Boston, another aging team with injury issues. I was surprised to see the Rangers among the AL base-running leaders, because even during their recent contending days they have made miserable mistakes on the bases. Texas was slightly above 0. I'm guessing the positive ranking could have been built on Stolen Base Runs -- with a trio of base stealers in Martin, Elvis Andrus and Alex Rios. 

There didn't appear to be much correlation between good base-running teams and teams high in the standings. Four teams currently in playoff position ranked in the top third in base-running, but three "playoff teams" ranked in the bottom third.