Outs Above Average and the St. Louis Cardinals Defense


Clockwise from top left - Tommy Edman, Paul Goldschmidt, Kolten Wong, Tyler O'Neill, and Harrison Bader

Earlier this offseason, Baseball Savant put up leaderboards for infield defense (how they determine it), which was previously uncharted territory at that site. I believe, after reading up extensively on DRS, UZR, and OAA, that Outs Above Average at Baseball Savant is clearly the best way to measure defense around the field, of course with caveats* because defense is so hard to quantify well.


*caveat #1 - OAA does not really yet exist for catchers. They have framing accounted for but nothing else *caveat #2 - OAA still does not (but will in iteration #2, according to Tom Tango) incorporate first base defense outside of fielding a ball - for example, scooping bad throws from other fielders - among other things (like it not having double play runs counted).


Those caveats included, I still believe OAA to be easily the best defensive metric out there in the public sphere due to everything I've read on all of the subjects.


(Note: Since starting this article, other metrics have begun to incorporate more into infield defense than their previous iterations as well.)

You're not here for that, though. You're here for information about the St. Louis Cardinals, of which I hope to provide you with plenty. First of all, I think everyone with half a brain (hi! that's me) and a pair of eyes (me again!) who watches the St. Louis Cardinals consistently could tell that their infield defense in 2019 was far superior to their infield defense in either 2018 or 2017. Baseball Savant has now put a number of runs on that improvement.


In 2017, the St. Louis Cardinals defense was a -4 in Infield OAA. That means, as a whole, the team was 4 runs worse than an average team on defense, just at the four infield spots. In 2018, they got MUCH worse, dropping to a -18, or 18 runs below an average infield defense. However, in 2019, that all changed.


The 2019 Cardinals were rated the best infield defense in the majors (after a lower-middle ranking in 2017 and a 3rd to last place finish in 2018) with a +42 runs saved above an average infield defense. That means from 2018 to 2019, the Cardinals infield defense got 60 runs better - approximately 6 wins better - just in their infield defense alone. THAT is incredible.


To further look into the St. Louis Cardinals defense, I needed to dig a bit deeper. I needed more information. Above I compared three seasons worth of data. That is what I have been led to believe is the right amount of innings to look at to get a representative sample on defense. Even doing so, at an individual player level, not every player on the Cardinals (in fact, very few) have this representative sample of three full seasons worth of data. Here is the raw data I collected. I like being transparent and showing everything I find when I can. I hope this is easily presented enough for everyone.

This is all of the data broken up by player AND by position (you can see these in the first two columns). In the 3-5, 6-8, and 9-11 columns, you should see the year, how many innings at that position, and the outs above average listed on Baseball Savant for that year in that amount of time. The last column is where I'm trying to get with all of the players. This is how many outs above average each player saved at each position over a 1,350 inning ("full season") sample size ON AVERAGE. They are listed from best to worst. Obviously, some of those numbers get quite wacky at the bottom with so few innings (like Molina being a 60+ runs worse than average first baseman given a season there. That's stupid and ridiculous and what happens with a 22 inning - less than 3 game - sample.)


What this chart should go to show you is that we have some very good defenders. Tommy Edman only has 600 innings anywhere on the diamond in the majors but is near the top of the list at two positions and near the middle of the list at the other two. He's a good defender. Even in that sample. Same with Harrison Bader. He's listed at CF and RF and LF, but in LF he's only played a little and played it average, but he was ELITE at the other two spots. Conversely, Dexter Fowler and Yairo Munoz** are at the other end of the spectrum. They, at multiple positions in the field, have been worse than an average fielder, but quite a bit at times.


**Yairo Munoz has since cost himself his potential role on the Cardinals by abandoning ship (per all reporting on the matter) and flying back to the Dominican Republic without alerting team officials, thus getting cut by the Cardinals in early March.


Where I would like to go next with this is to look at each of the Cardinals cumulatively. I combined how much they played each season at all positions. Here is that data.

Again, let's address the concern of sample size here. I don't believe that Rangel Ravelo will be a +30 defender (that's usually reserved for best season of the best player of a generation type of defense) in a full season. Also, like I said, no catcher defense, that means Yadier's numbers are just from first base in this sample. So please laugh at those.


Let's get into some great images by Nick Childress, the first being the cover art for this article. These are the players that are clearly above average in the field the last three seasons for the St. Louis Cardinals.

Clockwise from top left - Tommy Edman, Paul Goldschmidt, Kolten Wong, Tyler O'Neill, and Harrison Bader

The next fantastic image by Nick is the two guys who confuse people defensively. Neither of these left-side infielders are going to make the super-spectacular play to be on the highlight reels super often. But the Outs Above Average numbers show marked improvement in both players. Let's look at the last three years of Matt Carpenter and Paul DeJong.


Matt Carpenter went from a -1 to a -3 to a +6. That jump to a +6 came with slight defensive positioning adjustments (see this thread), but also with a full-time move to third base from a combination of first and third over the prior two seasons.


Paul DeJong went from a -7 to a -5 to a +12. While other systems (DRS/UZR) had DeJong as anywhere between average to above average each of his first three years in the majors, OAA meets more my eye test of DeJong. He really seems to be improving greatly year after year. OAA shows that for 2019 especially.

Paul DeJong and Matt Carpenter

Our last image is of two that are at the very bottom of the list - that have nott been good defenders no matter where they have played, as Cardinals, the last three seasons. When they don't hit, that makes for some really rough portions of seasons. You have to have something carrying your game. These two didn't / don't at times.

Yairo Munoz and Dexter Fowler

I hope this was a good (first for some of you) look at OAA and at the difference between it and some other systems. Hope you enjoyed. Glad to be back to getting some thoughts down "on paper" (so to speak) because I am having trouble talking baseball lately without anything going on. This is my release. Thank anyone who reads this to allow me to have this release from life at the moment. I miss you all and I miss MLB. Whenever it starts back up, LET'S GO CARDS!


Until then, don't be strangers.


(Thanks to Baseball Savant!)