Rethinking How We Look at Team Defense

They say that a run prevented is as important as a run scored, and this checks out. In fact, based on the coefficient of determination (r^2) for the two variables, a run prevented has actually been more correlated with team success than a run scored. This has indeed been labeled as the “run prevention era,” and just by that measure, this would appear to be the case.

As we’ve discovered in the past, offense and pitching wins championships, especially compared to defense. However, that certainly does not mean that defense does not matter. Rather, it is a small advantage that teams can leverage to continue to win between the margins. Small-market organizations such as Cleveland, the Rays, and the D-backs have all benefitted from strong defense in the past, while the Mets have been a clear example of what poor defense can do to you.

How can teams gain an edge defensively, and how much does it matter? What are the most important defensive positions, and how does it vary from the defensive spectrum? Should teams tailor their defense specifically to their pitching? Let us change the way we look at team defense by crunching through the numbers!

How Much Does Defense Matter?

The main principle of fielding independent pitching (FIP) is that pitchers can control strikeouts, walks allowed, and home runs allowed. Thus, the lower a team’s ERA is compared to a FIP, the better the defense had to be, right?

In general, teams with better defenses do outperform their FIP. This is almost entirely tied to being able to allow a lower batting average on balls in play (correlation coefficients are practically identical). Yet at the end of the day, defense still cannot completely predict outperforming one’s FIP, as only a third of the variation could be explained by it. Rather, their luck is often simply statistical noise.

To better explain this, think of a bloop single that scores two runs. Even the best defenders won’t be able to get to that ball, and in general, there are a lot of batted balls where defensive skill doesn’t have an effect on the outcome. Hence why good defense cannot have a complete direct effect on run prevention.

Should Teams Tailor Their Defense To Pitching/Positioning?

When the Mets acquired premier shortstop Francisco Lindor, many deemed him a perfect fit, as they have a lot of pitchers who induce a lot of ground balls.

This sparked a question: does infield defense matter more for ground-ball heavy pitchers, and is the same is true for outfield defense with fly-ball heavy pitchers? Surprisingly, not really. In fact, there was almost no difference between the effects of infield/outfield defense based on batted-ball tendencies. My guess is that while infield defense may matter more for a specific pitcher, no team has a full staff of pitchers of the exact prototype. Thus there isn’t any reason to be extra rigid defensively.

On the contrary, there is one way teams could tailor their defense to their pitching: choosing how much to prioritize it. Team defense did have a slightly greater effect for pitching staffs with a K/9 under 7.50, while it mattered less for those who struck out at least one batter per nine innings. This makes an incredible amount of sense; the more contact you induce, the more chances there are for a defense to make an impact for a game. On the other hand, a team like the Nationals, who have pitchers who strike out a lot of hitters, may not have to worry as much about defense, which would add up based on the additions of corner defenders Josh Bell and Kyle Schwarber this offseason.

Heading into this study, I was convinced that since more progressive organizations tend to utilize shifts more, there should be some correlation between having a low BABIP allowed and shifting a lot. However, this did not turn out to be the case at all:

To my surprise, more shifts doesn’t mean that you’ll allow a lower BABIP. At the same time though, the 2019 Astros and Dodgers each shifted at an extraordinary amount and a) outperformed their FIP by a considerable margin and b) allowed a low BABIP, so perhaps teams just haven’t been as aggressive at shifting in the past. Assuming shifts aren’t banned anytime soon, I am interested to see if shifting starts to make more of an impact. It certainly appeared to have an effect in the 2020 playoffs with the Rays, Dodgers, and Astros, though that is too small of a sample to make any substantial conclusion.

Which Positions Have The Greatest Defensive Impact?

Based on the defensive spectrum, here are the most important positions:

  1. Catcher
  2. Shortstop
  3. Center Field, Third Base, Second Base
  4. Corner Outfield
  5. First Base

The defensive spectrum factors in how difficult a position is, but when assessing team defense, that is irrelevant. Instead, we are simply focused on which positions actually matter when it comes to outperforming FIP and allowing a lower BABIP.

Let us start with catchers. Unlike other positions, they don’t really have an impact on batted-balls, but their contributions come more from stealing strikes. As it turns out, having a strong defensive catcher is quite important when it comes to run prevention:

Stealing strikes is so critical for pitchers. The difference between a 2-1 count and a 1-2 count, for example, is substantial:

*Photo via Quantam Sports Solutions

One stolen strike can make the difference in an at-bat. Thus not only is the defensive spectrum right to have catcher defense as exceptionally important, but we may be underrating catcher defense. It may be difficult to have a light-hitting player in your lineup, but it may be worth it at the catcher position, as long as the team has enough offense elsewhere. At the same time, this is what makes a tremendous all-around catcher such as Yasmani Grandal so valuable, and it certainly helps justify the Phillies investing in J.T. Realmuto, though the aging curve of catchers is another story.

In fact, catcher defense correlates greater with ERA than any position does with outperforming FIP, which is simply absurd. Speaking of which, here is how important each position is when it comes to run prevention:

As expected, first base defense doesn’t have much of an effect in general, so teams are right to continue to prioritize offense. Meanwhile, infield defense appears to be slightly more impactful than outfield defense, though they are essentially even in terms of significance.

Now for some of the surprises. What should we make of the diminished impact of second base defense? In my opinion, this correlates perfectly with where the game is heading. The Dodgers, for instance, have not prioritized having an everyday second baseman and have been okay playing Max Muncy there or moving immobile prospects with elite offensive skillsets to that position. The Twins, Yankees, and Rays have also been willing to move prospects with that profile to second base, and as we saw with the Reds signing Mike Moustakas as a second baseman, teams appear to be realizing this at the MLB level as well. This also might explain why the Cardinals were okay declining defensive-minded second baseman Kolten Wong’s $12.5 million club option. At the same time, the Giants felt comfortable handing out a three-year contract to Tommy La Stella, though defensive versatility may have played a role in that as well.

Additionally, the rise of third base defense has been tremendous to watch. Once seen as an offensive-minded position, it appears that third baseman are essentially a second shortstop, with Matt Chapman, Manny Machado, and Nolan Arenado among some of the premier defenders in the entire sport. Although shortstop prospects that may need to move off the position are riskier than those who will stick there, the risk with that profile may be overblown.

In the end, defense is key in the outfield (likely center field), catcher, and the left side of the infield. This fits tremendously with what positions organizations tend to place a known value on defense at, and it certainly reflect what positions cannot have a defensive liability hidden by strong defensive positioning. It will be interesting to see if teams continue to neglect second base defense, and also if third base defense is at its peak because of the defenders available currently.

Overview

What should we take away from this?

Yes, defense does help when it comes to outperforming FIP and allowing a lower BABIP, though the effects of defense are still limited compared to simple variance. For the most part, however, you shouldn’t tailor your defense to your pitching, though a whiff-heavy pitching staff would allow you to not focus on defense as much, and vise versa.

In terms of which positions matter most, the value of catcher defense cannot be overstated. With the automatic strike zone pending, that may not be the case for long, but for now, catchers are the only non-pitchers that have a direct impact on run prevention. Besides that, it is clear that second base defense is likely overvalued by the defensive spectrum, and teams are rightfully starting to prioritize offense at the position. Conversely, third base is becoming more of a defensive position, and can help a limited shortstop when we thought that the relationship only worked in one direction. Marcus Semien, for instance, improved his defense with an elite defender next to him in Chapman, and we’ll see if Chapman can have a similar impact for Elvis Andrus.

Defensive Runs Saved correlated more with run prevention than Defensive Runs Above Average, which aligns with Ultimate Zone Rating. Does this mean that DRS is the premier defensive metric? Not necessarily, but with fWAR taking into account DRAA, this is something worth noting. In general, however, the variability with defensive metrics shows the limits of evaluating players with one singular statistic, and in general, why adding up projected WAR isn’t necessarily a good idea. Defense at different positions impacts teams differently, and it is very difficult to project which teams will outperform their FIP.

The purpose of my personal value statistic, Value Over Average (VOA), is to project a player’s value in a neutral statistic, which is why it weights offense more than WAR does. Team VOA has been more correlated with team success than team WAR in the past, and I believe the limits of defense (it can’t have direct impact on run prevention) is why.

It also may be time to rethink the old defensive spectrum, as second base is clearly not on the same level as third base and center field. Not only is it less important, but with how many less mobile players are being put at the position, I’m guessing new studies could show less difficulty playing the position with an increase in shifts, particularly against left-handed hitters. At the same time, however, we are moving towards a position-less game anyway, which also may be a reason why teams continue to pay a premium for offense and versatility.

As always, one of the best parts about sports is that they continue to evolve. What happens to catcher defense with an automatic strike zone? What happens if the shift is banned? How many teams are realizing that the defensive spectrum may be outdated? It will fun to see these trends develop as teams focus on gaining small edges in any way imaginable!

More of my work can be found at mvpsportstalk.com.





newest oldest most voted
chefjj
Member
Member

Excellent piece, Justin. This really gets the brain’s gears turning when it comes to evaluating defense!

Jim
Member
Member
Jim

Very interesting.

abalasco
Member
abalasco

Great read. Properly evaluating framing seems so difficult, but the woba/count chart makes a very simple and obvious case for why it is important and how valuable that skill is.

jrroh
Member
jrroh

Really wonderful read. W/r/t the White Sox it brings up a number of interesting questions. They’re an old school team, who appear to have recently been turning away from the shift. Moncada at third grades out as a very good defender now, which seems to correspond to an improvement in Anderson’s defense at SS. That being said, Moncada’s defense at 2B was not great in his season and a half there. They were also the team who cut a popular gold glove second baseman because his arbitration salary was approaching $5m, foreshadowing what happened to Wong a year later.

jimfurtado
Member
Member
jimfurtado

I enjoyed the article but I have a question.

Why aren’t you using runs per game instead of ERA? If you are looking at defense, excluding runs scored based on defensive miscues doesn’t seem like a good choice.