Hello, and welcome to the pitcher edition of the projection summary for the 2018 Atlanta Braves. If you’re just stumbling across this post, I highly recommend you visit the position player post here first, because it has a lot more detail on what’s being discussed here, for the uninitiated. This post will look a lot like that post, but with a lot less stuff up front.
First, one question, the answer to which differs from that for position players.
Can you refresh me on the stats being analyzed?
Sure. Since I’m only really interested in aggregate performance, I only look at the following for pitchers.
- FIP is an “ERA estimator” that attempts to quantify how well a pitcher is pitching with a number made to resemble ERA. FIP itself only credits pitchers for strikeouts, and only takes away credit for walks and home runs allowed. Because I’m mostly interested in projection systems with regard to how well players perform as a way of assisting decision-making about how well teams are apt to perform, I very much prefer to focus on FIP rather than ERA, or xFIP.
- FIP- is basically wRC+, but for pitchers, and based on FIP. Roughly, an FIP- of 90 means that a pitcher’s FIP, after adjusting for park and league effects, was 10 percent better than league average. An FIP- of 120 means the pitcher’s FIP, after adjusting for park and league effects, was 20 percent worse than league average. As the run environment changes in crazy ways, FIP- has a lot more utility than FIP, because it’s hard to get used to the “new normal” for FIP (or at least it is for me). Note that since we don’t know what the run environment is going to be next year, any FIP- figures presented in this post are purely my own estimates, based on the 2017 run environment. To be clear, Steamer and ZiPS provide FIP for the Fangraphs player pages, but not FIP- (though Dan Szymborski has indeed provided FIP- in his team projection posts on Fangraphs). So, those estimates are purely my own, as are the inevitable issues associated with trying to forecast the future relationship between FIP and FIP-.
- fWAR and fWAR/200 IP are aggregate player production measures for pitchers. They are based on FIP, so you can see there’s kind of a theme happening here. One thing to note is that an FIP- of 100 will give you about 2.4 fWAR over 200 innings, rather than 2.0. This is just something to keep in mind for mental calibration, because it differs from fWAR/600 for position players, where being average across the board will yield 2.0 WAR, not 2.4.
Once again, I advise you to glance at the position player post here, if you haven’t already, for a discussion about the presentation of distributions of projections versus point estimates. Further, just like for the position players, IWAG models playing time for rotation members via a combination of injury and effectiveness. Some of the distributions are therefore kind of wacky: you may see some really long tails for fWAR/200, accumulated if the pitcher is pitching injured (or just having collapsed skill-wise) before being yanked, or if the pitcher has a great run and then gets injured.
I am once again replicating the following note, just in case someone does not visit the prior post: the dots for the Steamer/ZiPS projections on all charts below are placed there simply for illustrative convenience. I am not making any representation whatsoever that I have any knowledge of the Steamer/ZiPS outcome distributions, or that they look anything like the IWAG distributions being presented. Their placement is solely so you can compare all projections for a player on each chart.
Due to the relatively steady relationship between FIP- and fWAR, each pitcher really only has an fWAR and fWAR/200 distribution presented. The FIP- distribution would look mostly the same, anyway. I’m only adding a tiny bit of commentary for each of the 10 players examined. You can mostly use the distributions presented to see where IWAG is coming from and how that compares to Steamer/ZiPS, but of course I would be overjoyed to discuss any of this in the comments or via another medium.
For various reasons, Aaron Blair probably isn’t going to get much of a chance to start in Atlanta this year, unless things go all sorts of haywire. He’s had a strange development path and was worse in AAA last year than during his first go-around, but all the projection systems still see him as a #4/#5 type. IWAG’s overall spread is pretty wide — he’s probably something between slightly below replacement and league average in the end, but exactly where he ends up in there is anyone’s guess.
Mike Foltynewicz continues to be a contentious name in Braves Country, which is probably a matter of him not meeting the expectation that his arm and velocity could deliver an above-average pitching package. He still has the potential to do something akin to a league-average run prevention line or better, but he’s so inconsistent that you can also see a long stretch of ineffectiveness miring him in the sub-1 WAR quagmire. Given his inability to stay totally healthy, it’s hard to expect him to really clear 2 WAR by a sizable margin next year, but another repeat of some kind of mid-1.0s WAR is probably a safe expectation. The fact that his xFIP- actually went up from 2016 to 2017 is not necessarily a red flag, but like a burnt sienna flag, I guess.
Max Fried is an interesting guy. He has high prospect pedigree, but has lost a lot of time to injury and the results really haven’t been there since his return. He may have exciting strikeout potential buried somewhere, but for all the talk about fellow lefty Sean Newcomb’s walk rates, Fried’s are about as bad since his return from Tommy John Surgery. His first taste of major league action also wasn’t very exciting, and it’s a little unclear why the Braves called him up given that he had fewer of 100 innings of anything beyond A-ball at that point. One of my main questions for the universe is what Steamer sees in him right now, but we may never get that answered.
Of course, as indicated by the one-star confidence rating, this is all far more speculative on IWAG’s part than the distributions for other pitchers. We simply need more data to better get a handle on Fried’s future performance, though there’s not too much to really grab onto for optimism’s sake right now. Coming off TJS, Fried is also a bit more of an injury and playing time risk. Even if he does make the rotation, IWAG does not have the general expectation that he’ll crest 90 innings in the majors over the course of the season, at least not until the Braves shift to some kind of tandem or buddy system rotation plan.
Everybody Loves Luiz is probably a Brazilian knockoff sitcom, and an apt description of the general esteem in which projection systems hold Senhor Gohara. It’s quite interesting: IWAG is more optimistic than Steamer/ZiPS for position players, and is more in line with the other two systems for most (but not all) pitchers. Yet, Gohara, who is the 2018 Atlanta Braves rotation candidate that IWAG likes the most, has the point estimate below those of Steamer and ZiPS. (Gohara, though, is also by far the most preferred Braves pitcher of Steamer and ZiPS.)
IWAG’s general modeling for Gohara’s performance actually sees good groundwork for a 3.5-4 WAR/200 performance. But, he comes with elevated injury and adjustment risk, especially because he flew through the minors so fast last year. As a result, some of those drag him down, perhaps more than is warranted.
Also, you can pretty much ignore the long left tail. That’s really a modeling convention gone wacky that I didn’t bother to fix — while the likelihood of him posting -9ish WAR/200 is low, it’s essentially the result of modeling that sees him have 1 or 2 awful starts and then miss the rest of the season. In most distributions, those types of outcomes have such a low chance of occurring that they’re not even shown, but Gohara has so few career innings as a professional (fewer than 80 in 5 seasons) that essentially that tail didn’t get stripped out from the visual display. As with the case for this entire cadre of pitchers the Braves have assembled, the golden line is probably a better representation of what might actually happen with the pitcher, performance variation, injury, and all, rather than the green line.
Is Scott Kazmir healthy? Will Scott Kazmir get enough velocity back to stick in a rotation? The answers to these questions is a resounding maybe, or one of those ascii emoticons with the guy shrugging. If he does, he could be pretty good, as he has been in the past. If not, well, that’s what that left part of the golden distribution line is for.
McCarthy is kind of like a different shade of Kazmir. He also has some pretty severe injury risk, though his is more of the incidental catastrophic variety than the “will he/won’t he” recovery and questionable effectiveness variety. He’ll probably prevent runs at a league-average rate if and when he can pitch — if not do even better than that. But how often he’ll be on the mound to make that run prevention happen is a serious question.
Sean Newcomb is certainly unorthodox, given his sky-high walk rates, but that hasn’t stopped him from being effective. I’m not sure exactly what it says when all three projection systems think that your team’s best (Gohara) and second- or third-best pitcher (Newcomb) have a combined 130 major league innings under their belt, but that’s the kind of world that we Braves fans live in. Newcomb’s inefficiency on the mound is going to crimp his innings, so expecting 200 frames even if he’s a healthy workhorse doesn’t seem to reasonable, but IWAG doesn’t really see him running into an ineffectiveness buzzsaw. It might be kind of agonizing to watch, but he should continue to be relatively effective despite the walks and messy innings. (At least until hitters figure out they can just stop swinging.)
If you look at Sims’ stats, you kind of get why IWAG isn’t as high on him. He’s had dreadful walk rate issues in the minors, and didn’t really do anything particularly well when he made 10 starts and four relief appearances in Atlanta last season. My question is why Steamer and ZiPS see in him, but even so, it’s not a huge disparity, as he’s pretty much a tail-end rotation candidate either way.
Teheran also has the bimodal distribution going on, and Braves fans probably get that pretty well at this point. The right part of the curve is “good Teheran,” as we’ve seen in 2016. The left part is “bad Teheran,” as we’ve seen in 2015 and 2017. The fact that the left curve is bigger is concerning, but the point estimates are all in the middle of that. On the plus side, Teheran’s been durable, so he has a pretty good shot of providing whatever run prevention value he’s going to accumulate for much of the season. On the minus side, that just might not be a lot of value, though there’s a decent chance that it is. It’s just not the most likely outcome.
Wisler has a bit of the Lane Adams problem going on. Or, perhaps the Rio Ruiz problem. In short, Wisler has fallen flat in the majors so far, with just 0.5 fWAR in about 300 innings and an FIP- of 124. On the other hand, he’s routinely been better in AAA. So, the question remains: is he just doing something at AAA that plain doesn’t work at the major league level, or can those two different performance levels reach some sort of equilibrium where he’s a decent #4/#5? Anyway, I’m prepared for IWAG to be wrong on this one, and Wisler may not get much of a shot to make heads or tails of this projection anyway. It’s an interesting question, in theory, but probably not one that concerns the Braves too much.
A position-by-position summary doesn’t make much sense, but essentially, the rotation question for the Braves is one of health, durability, and depth.
- Luiz Gohara, Brandon McCarthy, and Sean Newcomb have the makings of a pretty good front end. But how many innings can you really rely on them for? They might get 300-310 altogether as a central estimate, which is about half of what you’d expect three very durable and quality pitchers to give you. So, you basically need another 3.5 full-time rotation spots.
- Teheran can cover one, but may not be very good in doing so. Foltynewicz can give you three-fourths of another. That leaves 1.75 rotation slots, and the field opens up.
- Kazmir could be decent, but again, how many innings can he really be relied on to provide?
- After that, you’ve got Blair, Fried, Sims, and Wisler, and you’re going to need them to combine for over one full rotation spot’s worth of innings. Is that something likely to deliver decent value?
Overall, the IWAG point estimate for the rotation is about 9ish WAR. The Fangraphs Depth Charts (which are an average of Steamer and ZiPS, but use different playing time estimates) have the team at around 10 rotation wins. Those are both midrange numbers, but nothing special. The more high-quality innings that Gohara and McCarthy are able to provide, the better, but the lack of average-type depth is what might really hurt the Braves if injuries strike.
Again, just for fun, there are projections below for Kolby Allard, Wes Parsons, and Mike Soroka, each of whom might pitch in some starts for the team in 2018. Just like Austin Riley in the position player post, the IWAG projections here are best interpreted as “if things keep going the way they have” projections and not “if they made the majors right now” projections.
The Wes Parsons projection really surprised me, because he is not really talked about as a top prospect, or anything close. In a swingman role, though, Parsons has had a lot of success in High-A and AA over the last two seasons. Maybe that fun train will stop when he gets to AAA, but on a results basis, he seems to be interesting enough to not drop off the Braves’ radar. Of course, this could also just be another case of Four-A or “improper skillset alignment” syndrome, where whatever Parsons is doing simply won’t work in the majors. But I’m now interested enough to keep an eye on him, whereas he was essentially a non-thought before.
Up next, relievers — the bane of projections, and prognostication in general.