What is the Cardinals best outfield?

Listen. I’m tired about talking about the outfield situation. You’re tired of hearing about it. You’re tired of thinking about it. Any discussion or thought about the Cardinals inevitably leads to a discussion about the outfield. The is the topic of the 2020 season, and to be honest, it looked like it would be that way before the season began. I would love to talk about something else.

But

But there’s a reason for all that. I can’t NOT talk about the outfield situation. I’m sorry. The pitching situation is basically “Is your arm still attached to your body?” and “If yes, you can pitch” and “If no, you might still need to pitch.” The infield and DH spot are more or less solved and will start every game they can. The outfield though, oh god the outfield.

Here’s what I want to do. I want to strip the entire conversation to its basic parts: who are the best players? That’s it. I’ll use projection systems - ZiPS and Steamer combined - and I’ll personally estimate their defense based on past results and speed and a little bit of eye because some of these guys have no sample size at all. The bigger the sample, the more I’ll use past results. I’ll explain my defensive estimates the best I can.

Important note: there’s no way my WAR numbers are exactly right. I used the same site for every player though, so they should be correct in the sense that you can judge them against each other, which is really what this exercise is about.

Dexter Fowler

I realize he’s hurt and out of the equation, but let’s start with him so I can give you a good idea of how I’m doing this. Dexter’s projected slash line for the rest of the season, combining ZiPS and Steamer, is .238/.337/.406, for a 99 wRC+. For his defense, I used his last 3,900 innings, dating back to his centerfield days in Chicago. In those innings, he has been the equivalent of a -3 defender, although a lot of that is boosted from his 2016 numbers at Chicago, and he’s 34 now, so I feel pretty comfortable calling him a -5 fielder in RF. He’s an average baserunner. So we can expect Dexter, over 600 PAs, to be a 0.9 fWAR player.

Harrison Bader

Nobody else will be as easy to project as Fowler. Harrison Bader has just over 2,000 innings in the big leagues, which is about 1,500 innings short of a good sample. He has the largest sample. You can see the difficulty I face with projecting defense. The default assumption when a player doesn’t have enough innings is to assume the rest of his innings played are a net zero. I don’t think we have much reason to do that for Bader, given his immense speed and that ole eye test. Looking at the career defensive leaders since 2010, several elite defenders hover around the +10 mark, so I feel that’s a reasonable assumption for Bader for those remaining innings.

Bader in those 2,100 innings, has been the equivalent of a +17 fielder (his RF numbers do not really help him). Give him +10 for the remaining innings necessary to have a good sample, and he’s a +13.9 fielder. Kevin Kiermaier, for his career, has +15.2, so this seems like a very doable number for Bader, especially with his head start. Bader, according to Steamer, is roughly a +2.3 baserunner over a full season. By my calculations, Bader is projected to be a 3.7 WAR player over 600 PAs. That is quite good.

Now, that’s the rest of the season projections based on his hot start. This projection is certainly overrating him. For instance, it’s probably putting a bit too much on the fact that he has a .261 ISO for the moment. Bader doesn’t appear like too much of a different hitter than his original projection, which was a 93 wRC+. The updated is 97. Which makes a difference but he’s not as far off from 3.7 WAR as you’d think. Seems like 3+ WAR player is a decent bet.

Tyler O’Neill

Now here’s an interesting case. Because his numbers this year have not been very good. But he shows improvement in other areas. Nevertheless, he has taken quite a hit in rest of season projections. ZiPS and Steamer are in lockstep over his offensive production, which they see as a .230/.299/.434 hitter for a 92 wRC+. That’s a 5 point drop in wRC+. Nearly all of it is BABIP and to be kind of fair, I think rest of season BABIP ends up getting more affected than pre-season. Which is to say, I don’t think ZiPS is projecting a .280 BABIP for him in the 2021 season, especially given his career BABIP is still .332.

Now his defense. Hmm. Well the early returns are good, albeit nearly useless. In less than 1,000 innings at the corner outfield positions, he’s a +10 fielder. He’s also fast. But he’s awkward as hell out there. But LFs traditionally are garbage fielders so his competition isn’t great either. I’m just going to take a stab in the dark here and call him a +5 fielder in LF. You can be okay with it or not. Ultimately, his speed plus LF being the position of the Manny Ramirezes of baseball makes me think he’s rather likely to be above average, but his awkwardness makes me not want to go higher than +5.

I will also assume he’s a + baserunner. Sample isn’t large, but he’s been +2.4 with barely over half a season and again, his sprint speed is near the top of the league. So the improved plate discipline Tyler, with some assumptions on baserunning and speed factored in, happens to be a 1.9 WAR player over 600 PAs. So he’s currently projected as a below average player.

Dylan Carlson

I have no idea how I’m going to do defense here. Change his WAR based on whatever you think his defense is. We’ll start with his speed. Speed isn’t everything with defense, but it certainly helps. Carlson has a sprint speed of 27.4 feet per second so far. It’s too small of a sample I think, that would have ranked 237th in the league last year out of 568 players. It would have tied for the 8th slowest among 75 players Statcast called a centerfielder. I don’t think he’s a major league centerfielder.

Statcast has him as -1% success rate added in CF, -8% in LF, and 2% in RF. None of this really means anything to be honest. Given his speed and yes, my eye test, I don’t think he’s an above fielder in center. But he looks pretty good in the corners. I’m just going to go with +5 in the corners. At some point, you just have to guess on these things. I don’t think the +5 positional change would work on Carlson, so I do not think he’s a dead average defensive centerfielder. But I’m going to assume we mostly play him in the corners for this projected WAR.

His Steamer/ZiPS projection for the rest of the season is .239/.309/.404 for an 88 wRC+. I don’t know how much Fangraphs wOBA and Statcast wOBA align, but his xWOBA is just .283 this year, and his projected wOBA is .303. I don’t really believe in xWOBA granted - too many players have considerably different xWOBAs over large samples for me to believe in a small sample one. And I’m going to assume he’s a neutral baserunner. His BSR is shockingly bad at -1.6 already, but the sample is much too small. Carlson is projected for 1.3 WAR over 600 PAs based on rest of season projections.

Lane Thomas

Lane Thomas is in a similar spot to Carlson on defense, but he’s fast. He doesn’t have enough of a sample for him to even show up on the 2020 leaderboard, but he was 15th in sprint speed in all of baseball. However, I don’t want to read too much into that sprint speed. It’s based on 21 competitive runs. Plus, 15 of the top 30 in 2019 were centerfielders. I assure you not all of them were great on defense. So I am forced to throw out a guess, and accept it or not, I’m making him a +5 defender in CF. You can’t really assume better than that with literally no sample.

On offense, his combined projection is .232/.303/.391 for an 84 wRC+. I’m a little higher on him than that, but I don’t think his below average line in Memphis helped much here. I’ll assume he’s a + baserunner. Over a full season, he is higher than I expected, although you can really see why Bader has as high of a projected WAR as he does when you know how valuable simply being a plus defender in center brings. His projected WAR is 2. This seems way too reliant on baserunning and defense for a guy with virtually no track record so that seems extremely optimistic. But there you have it.

Tommy Edman

The question isn’t really how good of an outfielder Edman is. The question is how much value you’re removing from Edman’s game by putting him in the outfield instead of the infield. I know Edman’s fast, but he’s a below average defensive outfielder at the moment. I will however assume he’s an average defender for this post. Him being average defensive OF is best case with how he’s looked, so call this choice a curiosity.

His combined projections are .269/.321/.402 for a 92 wRC+. Assuming he’s merely an average left fielder right now - which is a generous assumption in this writer’s opinion - Edman as a plus baserunner is projected as a 1.3 WAR player in LF. Which tells us something we all know: he’s the worst choice to play the outfield of this group.

But let’s compare this to him in the infield. Assuming he’s an average 3B, and I think he’s probably better than that, he’s a 2.3 WAR 3B. You’re literally losing a full win, probably more, by playing him in the OF instead of in the infield. For whatever it’s worth, Matt Carpenter with a 105 projected wRC+ and -5 defense at 3B... is also 2.3 WAR. But I imagine most people, myself included, would take the under on Carpenter’s offensive line. It requires some semblance of power, and he has shown basically no signs of having that.

Conclusion

Technically speaking, the best outfield alignment is probably Harrison Bader in CF, Tyler O’Neill and Lane Thomas in the corners. Of course, Thomas in RF with an 84 wRC+ would require Jason Heyward levels of defense, so that really wouldn’t work, which would mean Carlson is probably a better option. So there you have it. Bader would pretty much start in CF every day for my team and then the other three would rotate until another emerges. That would be my plan.