Who is the ideal internal replacement for the third outfield spot?

With two of the three planned starting outfielders on the injured list for the time being, the Cardinals have a much more uncertain starting lineup than how the season began. Two of the three are easy enough: Dylan Carlson has both the performance and prospect pedigree for a start. Justin Williams doesn’t really have either, but what he does have is some sort of potential ability to start in the future. The rest of the outfield doesn’t really, so he’s a starter by default. And that’s it.

So I wondered. What’s the optimal starting lineup right now then? The Cardinals appear to think of Austin Dean as a platoon player - against lefties - and John Nogowksi as not an outfielder at all. Lane Thomas can’t be regarded as a potential starter, or, well he would have made the team out of spring training. Which is why you’re seeing Matt Carpenter start at 2B and Tommy Edman at RF. Are they right?

Let’s run some numbers. But first, I have the annoying task of having to mostly make educated guesses at some defensive ability here. We pretty much know Matt Carpenter’s defense, although him not playing 2B much in a few years complicates even that. We have a reasonable idea on Austin Dean as well. And the rest.... I mean I’m not pulling these guesses out of thin air, but I’m not working with much here. I’m just gonna have to make some assumptions.

Let’s start with the easy one, Carpenter. At 3B, he is a career -3.6 UZR/150 fielder and he has been roughly a -3 fielder the last three years. At 2B, he’s a career -5.8 fielder at 2B. 2B and 3B defense is on the same level of the defensive spectrum, so the assumption is that players are roughly the same at each. That said, I’m willing to take that 2B number at face value for the sake of simplicity. You’d imagine he’d be worse at 2B just for less experience.

Austin Dean is also not that complicated. The only trouble is it’s a small sample, but the Jose Martinez comparison is not without reason. Back when Martinez was on the team, I figured, long-term, as bad as he could be, using historically bad defenders, was about -10. That’s kind of what we’re working with on Dean. The assumption with this kind of thing is to not go worst case, so in that vein, I’ll just say he’s a -8 fielder. That’s still very bad to be clear.

John Nogowski actually isn’t hard either. He has basically no experience in the outfield. If the Cardinals thought he could play outfield at all, he’d probably have gotten a start out there by now. Worse for him is his sprint speed. He doesn’t have the sample to go on the leaderboard, but his 24.7 ft/sec sprint speed would rank 156th out of 167 qualified hitters. He is very slow. There is no reason to think he’d be anything but a disaster defensively in the outfield. Since he’s not an actual OFer and he’s slow, I actually think he’d be below my worst case number above, because that number was based on OFers who accumulated 3,000 innings, hence actual outfielders. So I think Nogowski would be something like a -12 defensive outfielder.

Edman is tricky. Because there are times when it’s very clear he’s inexperienced defensively and yet there are times when he also seems fine out there. He’s also really, really fast. Speed makes up for a lot. His early Statcast numbers are promising. Since 2019 as a RF, he has supposedly been a +2 fielder at RF. That’s a pretty impressive number for a guy who never played OF and who is working with a grand total of less than 50 total attempts. (I think OAA is a counting stat). It’s very hard for me to not come to the conclusion that Edman isn’t probably at least average in RF. And my personal guess is that he’s a little above average there, barely. I’ll say +1 in RF.

Lane Thomas is, I think fairly clearly below average in CF, as yesterday made clear. But I will say I’d find it rather hard to believe his speed alone wouldn’t make him above average in the corners. Some of his mistakes yesterday were of the lazy variety, which speed can’t really cure. But his early defensive numbers are positive, if almost completely useless with the sample. I’m going to say he’s +3 in the corners. I think there’s a decent chance there’s something about him specifically that makes him worse at CF than you’d think, but I think in the corners, I’d struggle to imagine him below average there personally.

So Thomas is a +3 in the corners, Edman a +1, Dean a -8, and Nogowski a -12. Plus, Carpenter is a -6 at 2B. And since the alternative to Carpenter at 2B is Edman, I must give him a number at 2B. I’m going with +5 defender at 2B. He’s already +2 this year at 2B according to Statcast and he was once +6 at 3B and +2 at 2B in the same year, so I think he has even more potential than that. So +5 seems like a safer guess.

I understand there are some Nogfather believers out there, but I just feel like it takes too much imagination to think he’s a good option in the OF. If he’s even -5 in the corners, which feels highly optimistic in my opinion, he’s still just a 0.5 WAR player per 600 PAs. Anyway, I can at least fairly easily conclude that neither Nogowski nor Lane Thomas are a part of my ideal lineup right now. Which puts the discussion at Dean/Edman or Edman/Carpenter.

And the answer, for the moment, appears to be Dean and Edman, entirely because Edman is much more valuable at 2B. Due to the positional value differences between 2B and RF, Edman would need to be a vastly better fielder in RF than he is at 2B and there’s simply no way. So with 2.9 WAR to Edman/Carpenter’s 2.4 WAR, it appears that’s the answer.

However, the Cardinals are treating Austin Dean like a platoon player, which seems to indicate they think he’s worth playing against LHP, but not RHP. Interestingly, Dean has been better against RHP in his short MLB career so far with an 89 wRC+ against RHP, 60 wRC+ against LHP. But do not read into this too much. We’re talking a grand total of 104 PAs.

Plus, I checked his minor league splits. In 776 PAs spent mostly in AAA in 2018 and 2019, he had a .819 OPS against RHP. In 266 PAs in that same time span, he had a .917 OPS against LHP. In 616 PAs spent in AA between 2016 and 2017, Dean had .691 OPS against RHP. In 188 PAs against LHP, he had a .849 OPS against LHP. In 2015 in High A, it was .644 OPS against RHP, .789 against LHP. In 2014 in A ball, it was .805 to .893. A little over 100 PAs does not trump that minor league data nor how the history of splits work, in that reverse splits are pretty rare.

So Dean is extraordinarily likely to be worse against RHP, despite what his current MLB splits tell you. If Dean is your average right-handed hitter, you’d expect his wOBA to be about 7 points worse against RHP than his overall wOBA (16 points better against LHP). Which would make him roughly a 0.3 WAR player against RHP. Then it’s 2.6 WAR versus 2.4 WAR. And well, you’d expect Matt Carpenter to be better against RHP than his projection as well.

But I’m not going to get into that. I know a lot of you have trouble buying into Matt Carpenter’s projection as it is. All I will say though is that, given the lack of any real good options, Matt Carpenter at 2B and Tommy Edman at RF is very defendable. Now that says more about the other options than saying that’s a good option, but until O’Neill returns, worst case we let Carpenter fail now while there are not better options, I say.