clock menu more-arrow no yes mobile

Filed under:

How the Cardinals Rank in Infield OAA

MLB expanded their new stat - Outs Above Average - to include infielders. Where do the Cardinals rank?

Divisional Series - Atlanta Braves v St Louis Cardinals - Game Three Photo by Scott Kane/Getty Images

Measuring defense has always been a challenge for major league front offices, coaches and fans. A simple stat, like errors, is incapable of considering all the nuance involved in making or not making a defensive play.

I was first exposed to this when Edgar Renteria was acquired from the Marlins heading into the ‘99 season. Renteria arrived with the reputation of an agile and promising young defender. Over the next three years, he would commit 77 errors, the most in baseball. Those high error totals sparked an on-going debate. One side claimed, emphatically, “Renteria can’t be a good fielding shortstop! Look at all the errors!” Others rebutted, “He’s only making errors because he’s got so much more range than other shortstops!” Who was right? “Me”, of course, no matter which side “me” was on because there was no fielding metric that could objectively answer the question.

Through 2009, Cubs shortstop Ryan Theriot had a high of just 15 errors in a season - relatively low totals for a shortstop. In 2010, the Cubs became dissatisfied with Theriot’s fielding and moved him to second base to make space for rookie Starlin Castro, who was no defensive wizard. The same year, the CardinalsBrendan Ryan was fielding (and, unfortunately, hitting) like a young Ozzie Smith. After the season, Ryan was LaRussaed (yes, that’s a verb) to the Mariners and Theriot was brought in as the answer at short. The club cited the veteran infielder’s low career error totals as one reason they believed he could move back to short and succeed there. Let’s just say that the move didn’t work out quite as hoped...

The point is that whether we are talking about a great defender, like Renteria or Ryan, or a poor one, like Theriot, traditional metrics routinely fail because the stats cannot consistently account for all of the variables that affect a single defensive play.

That’s the idea behind MLB’s new infield OAA — Outs Above Average. Mike Petriello of MLB.com provided a nice summary of what OAA is trying to accomplish (you can find the introductory article here):

There are four primary items that affect the chance of a play being converted into an out:
• How far the fielder has to go to reach the ball (“the intercept point”)
• How much time he has to get there
• How far he then is from the base the runner is heading to
• On force plays, how fast the batter is, on average

Infield Outs Above Average attempts to factor all of these variables into every ball in play on the diamond. It considers the position of the fielder, the exit velocity of the struck ball, the distance the fielder has to travel to field the ball, and even the average sprint speed of the baserunner. All of this information is made available through Statcast. OAA is unique among advanced defensive metrics because it can also factor in shifts, calculating data based upon the actual starting position of the fielder, regardless of where on the field that starting position is located.

OAA plugs all of that statistical soup into some kind of digital defensive Instant Pot and out comes a value that represents the percentage above or below each play as compared to an average out.

For example, if Paul DeJong fields a ball that’s hit relatively close to him, at a relatively average speed, by a relatively average baserunner, and he makes the play, then he gets 0 outs above average. He was supposed to make the play. He did make the play. That’s expected.

But, what about the more difficult plays? Like these:

Baseball’s new formula can determine with some level of confidence exactly what percentage above or below average each of these plays are. Add up all those individual plays over the course of a season (or three) and MLB can tell tell us, with just about every conceivable factor included, whether an infielder is elite, above average, pretty bad, or straight-up Ryan Theriot.

So, what does OAA have to say about Cardinals infielders? Instead of re-compiling the data here by player, I’ll just provide the team page as it is at Baseball Savant so you can see the fantastic way that MLB is presenting the data:

Take note of the compass-like graphic for each player. These graphics illustrate each player’s defensive OAA according to four quadrants: plays forward, backward, left and right. Paul Goldschmidt, for example, is pretty decent in every defensive direction. DeJong, however, is second best in the game going toward first base (his glove side) but is not so good going across his body toward third base. This seems to be the norm for shortstops. (Only Javier Baez was greater than +2 going toward third). Matt Carpenter, then, appears to have been a surprisingly good complement to DeJong in ‘19. Carpenter excelled moving into the range where DeJong struggled. (Pro-tip: Look up Nolan Arenado...)

On individual player pages, Baseball Savant provides a breakdown of a player’s fielding by both starting and ending location. I’ve combined DeJong’s two charts here:

Overall, Cardinals infielders ranked first in combined OAA, providing +42 Outs Above Average, way ahead of the second place Rockies at +33. The Padres were the worst at -23.

The individual rankings within the Cardinals largely make sense. DeJong (7th) and Wong (13th) are among the top fielders in the game according to OAA. This agrees strongly with more established defensive metrics like DRS and UZR. (DeJong ranks 11th in DRS and 4th in UZR. Wong ranks 12th by DRS and 21st by UZR.)

OAA, DRS, and UZR are not necessarily interchangeable. DRS and UZR are both run-oriented. Neither of those metrics include the same level of detail as OAA and largely make no effort to account for shifting, which is increasingly common. Still, because all three metrics are on a +/- scale, they can be used in coordination to provide a more complete picture of a fielder’s ability — similar to how we might use ERA, SIERA and FIP to evaluate a pitcher.

Carpenter’s evaluation as a +6 third baseman is the most surprising among the Cardinals’ infielders. That ranks 7th in baseball at the hot corner. Is Carpenter the 7th best third baseman in the league? Was Ryan Theriot a good SS just because one stat (errors) said so? That’s where I can offer a word of caution: defensive stats, like burritos, are volatile and best consumed in bulk (large sample sizes and multiple flavors).

In 2018, Carpenter’s OAA at 3b was -2, 22nd in the league. He was not good going to his left in ‘18. DRS and UZR consistently point to average or below defense from Carpenter at third throughout his career. Don’t assume that one positive year of OAA means Carpenter is or will remain one of the better third baseman in the league.

If this kind of information is your thing, Tom Tango provided a virtual dissertation on the subject and MLB has made that available to the public. Have at it! OAA adjustments will be updated daily throughout the season, so we should be able to track it as we go in 2020 and tie changes in data to specific plays in real-time. I plan to do monthly OAA updates in-season, with plenty of video highlights.