clock menu more-arrow no yes

Filed under:

How many games would an all-defense team win?

New, 8 comments

Not all elite defenders are elite players, but a team with the most elite defender at every position would still win a lot of baseball games.

Kevin C. Cox

We live in a baseball age where the value of defense is much more appreciated -- but is it appreciated enough? As we look at that question, let's assemble a team of defensive all-stars and try to determine how many games they'd win in a series. To stick as close to the real world as possible, we'll use actual players, based on 2013 statistics.

So here are your 2013 Defense-First All Stars, per Ultimate Zone Rating (since we'll be using FanGraphs WAR):

Position Player UZR WAR
1B Mike Napoli 9.7 3.9
2B Darwin Barney 12.5 0.4
3B Manny Machado 31.2 6.2
SS Andrelton Simmons 23.9 4.6
LF David Murphy 10.8 0.4
CF Carlos Gomez 24.4 7.6
RF Gerardo Parra 26.6 4.5
C Russell Martin -- 4.1

Filling out the squad

Before we rubber stamp our new team, a few other matters: outfielders, catchers, and pitchers.

I won't disturb the WAR total arrived at with a Murphy-Gomez-Parra outfield, but we can acknowledge that players are probably more able to move around in the outfield than infielders in the infield. There may be plenty of center fielders who, in 2013, could have moved to left field and put up superior UZR and WAR totals. A defense-first team might make that kind of move, but while Parra's mold-breaking tenure might cause us to wonder if teams may start to treat right field as a skill position, it's really hard to see a team committing to a Darwin Barney equivalent in left.

Catchers are a little more complicated, as we know. But in addition to adding value with good fielding plays and limiting the running game, Russell Martin was also a top notch receiver. At FanGraphs, runs added by skill in blocking pitches is calculated (RPP) and incorporated into WAR. Recent work by Dan Brooks and Harry Pavlidis suggests that the run value of blocking pitches can go beyond the +/- 7 runs in the model used by FanGraphs (statistics at Baseball Prospectus show 14.7 runs of value for blocking by Martin for last season), but that may not be a major point. We can account for framing, though, and include 19.2 runs worth of framing value for Martin (per BP) in his WAR total, knocking it up to 6.2 WAR using a nine runs per win framework.

As for pitchers, it would be unrealistic to think that any team would prioritize pitcher defense (even if one seems to have prioritized pitcher hitting), but it's not unrealistic to think that a defense-first team, like the one we're assembling, would at least work to train pitchers on defense. As a proxy, we could pick out the top defensive pitching staff of the 2013 season per Defensive Runs Saved: the Atlanta Braves, who notched 15. That's barely 2x the total put up by last year's top pitcher, Patrick Corbin (8 DRS), but good luck putting together a staff of five Corbins with a bullpen full of Corbin-like relievers (and good luck filling out your staff for 2014 if you did). For our exercise, however, we're going to have to go in a different direction, as there's no way to translate DRS (keyed to league average) to wins above replacement.

Considering the average position-player WAR for a pitching staff is 0.1, I'm not sure this makes much of a difference if our Defense-First All Stars team is in the NL. If you wanted to add a DH bonus for our AL team, be my guest, but that gets into some icky issues about how WAR might translate to wins in the real world.

How many wins?

Remember, this is about the number of games a team of best fielders would win. I'm not saying you could assemble these eight guys (and pitching staff) and magically win a certain number of games -- because defense is highly volatile, and because these eight men are all outliers, we can safely expect that their defensive marks from 2013 were a combination of skill and luck. But having concluded that a team full of Gerardo Parra-equivalents might win about 97 games, even with a so-so pitching staff, I'm curious as to how the fictional exercise would work without a fictional cloning machine.

Peace in our time between FanGraphs and Baseball-Reference for replacement level win rate (29.4%) means we can start with a baseline of 47.6 wins. The total WAR for our eight players above is 33.8. Even with a replacement-level pitching staff, our team of Defense-First All Stars may be in line for 81.4 wins.

With all due respect to the 2013 Houston Astros, however, I think it's fair to say that like any team, the Defense-First All Stars wouldn't be satisfied with a replacement-level pitching staff. Ignoring those 1.6-WAR 'Stros, the worst pitching staff in baseball last season was that of the San Diego Padres (4.4 WAR), while the Tigers featured the best (29.3 WAR).

The Defense-First All Stars, therefore, might finish as low as 85.8 wins (Padres staff) or as high as 110.7 wins. With a league-average pitching staff valued at 14.3 WAR, the Defense-First All Stars might finish around 95.7 wins -- they'd be a playoff team.

Final thoughts

Asking you to suspend disbelief for this exercise was asking a lot, I know. And there are flaws, even for what it is -- a fair amount of luck shows up in our WAR total because we picked only outliers, and some defensive prowess may be double-counted here, as there are some stupendous plays that, for example, Manny Machado and Andrelton Simmons can't both make.

There's no organization in baseball that would consider putting together a team of Darwin Barney equivalents (65.1 wins with a league-average pitching staff), which led me to believe that this inquiry would be best completed using the players who happened to be the best fielders in 2013. But some teams do play Darwin Barneys, and it's not always because the team is keeping its options open for uber-prospects that might have to move off of the shortstop position.

No, sometimes teams play a guy like Darwin Barney because he's cheap. Before he climbs the arbitration ladder, a player like Barney could allow you to divert resources elsewhere, and with market prices for defense-first players perhaps not where they should be, I think we'd find that most attempted Defense-First All Stars teams would have good money for other upgrades.

As it so happens, our Defense-First All Stars cost just $33.48M for 33.8 WAR (crediting Mike Napoli with the full $13M actually earned). Yes, we happen to have some league-minimum players in the group (Machado, Simmons, Barney), and some were arbitration cases (Parra, Murphy, Gomez). But it still stands to reason that a team that prioritizes defense in a big way could buy a pretty decent pitching staff -- and win a lot of games.

. . .

All statistics courtesy of FanGraphs and Baseball Prospectus.

Ryan P. Morrison is a featured writer at Beyond the Box Score, and co-author of Inside the 'Zona, a site on the Arizona Diamondbacks with a sabermetrics slant. You can follow him on Twitter: @InsidetheZona.