I’ll level with you, FanGraphs reader: I can’t imagine that FanGraphs analysis is exactly what you’re most interested in right now, baseball-wise. The exciting things going on are the games on the field as they happen. These playoff races are amazing. The schedule sets up with a wonderful mix of both-teams-in games and rivals playing spoiler. In that context, I can’t imagine that many people are dying to read about Tommy Kahnle’s changeup-heavy arsenal, just to pick on a random article I recently wrote.
That’s never stopped me from writing about whatever random thing caught my eye, but I thought I’d take advantage of a slow week (again, just in my kind of baseball analysis, not in actual baseball) to go over some predictions I made before the season. I like to look back on my own work for a few reasons — not just to bask in successes, but that’s certainly part of it. It’s also useful to talk process and separate bad outcomes from bad ex-ante decisions, though to be honest, that’s really hard, so I’m not certain I’ll get it right today.
I’ve named the column after one of my favorite weird sports things. I always knew this, but I’ve noticed it more since I started playing in a recreational softball league. After you return to the dugout, there’s a pretty good chance someone will slap your butt with their glove. It doesn’t matter what you did, there’s always a reason to. Did you score a run or make a great defensive play? It often comes with an “attaboy” or “way to go.” Did you make an out or embarrass yourself in the field? “Get ‘em next time” is nearly guaranteed. The butt slap? That’s a constant. The words that go with it? They’re versatile. Without further ado, let’s figure out whether my hypothetical dugout should be congratulating or consoling me.
I wrote a mini-article about four players I loved for 2023 after a panel discussion at the SABR Analytics Conference. You’ve probably heard me wax poetic about the first player on that list: Lars Nootbaar. I don’t think this was a particularly bold claim; as I mentioned in that article, he really broke out in 2022, and I was basically just speculating that the things he did well last year would continue into 2023.
Whether you’d score this as one in my column comes down to how you feel about merely repeating last season. Nootbaar posted basically the same wOBA and wRC+ as last year, but with more BABIP and less power. His power still looks good, to be clear; whether you’re interested in Statcast numbers or just the eye test, it looks about the same as last year, just with less home run luck. He posted the third-best batting line among Cardinals with reasonable playing time and racked up the second-most WAR despite missing a handful of games throughout the year.
I can’t give myself much credit for saying that a thing that was already going on would keep happening. But to the extent that one of the skills in making breakout picks is finding breakouts after they’ve happened and glitzing it up to make the pick feel more fun, I think I did an excellent job here.
My other hitter breakout in that article was Riley Greene, and he’s a more interesting case. I thought his strikeout rate was bound to improve, even if he didn’t change his approach at the plate, and probably even if he started swinging and missing more. Hey, what do you know: his swinging-strike rate increased by more than two percentage points, and yet his strikeout rate declined marginally.
The hellacious power he displayed in the minors was another reason to bet on Greene, and that also played out. He absolutely clubbed the ball this year and deserved better power output than he got even as he increased his ISO by 50 points. Sure, he had his fair share of batted ball luck when it came to hitting for average, and he had Tommy John surgery that ended his season prematurely, but the thesis that Greene would strike out less and do more damage on contact even without meaningful changes was pretty much spot on.
These ones weren’t so good! In that same mini-article, I highlighted two pitchers: Taylor Rogers and Tyler Anderson. I’ll start with Rogers, who was mainly in that article because I needed four names to talk about and only had three I really liked. My thesis there was that he had a high floor as a solid lefty-killing reliever, and that if he developed a cutter, he might turn into one of the best overall relievers in the game.
That didn’t happen. Rogers is still the same sinker/slider pitcher he’s always been, and he walked more batters than ever before in his career this year. Of particular worry, he walked 10.9% of opposing lefties; I can tolerate high walk rates from platoon specialists as long as they’re walking the tough matchups, but he lost winnable battles too often.
If there’s a lesson here, it’s not betting on a pitcher to come up with a new pitch. I don’t feel particularly bad about thinking Rogers had a high floor; his walk rate ballooned and he still posted a 3.71 FIP and 3.33 ERA. But a breakout implies something exciting, and he doesn’t really fit that bill, so this one seems like a miss, though not a particularly damaging one.
Anderson was a lot worse than I expected. He, too, had serious command issues; his 10.2% walk rate was more than double last year’s mark. I hypothesized that his new drop-down sinker would help him handle same-handed hitters, a strange problem for a junkballing lefty, so of course he got absolutely tattooed by lefties this year, and that sinker seemed to be part of the problem.
The thing that made me interested in Anderson in the first place was his arm slot change against lefties, and he doubled down on it this year. He added a slider from that lower arm slot, though he didn’t use it particularly often. His biggest issue was that he couldn’t command pitches from that low release point very well; he walked 14.3% of opposing lefties, an absolutely awful mark, and that goes a long way towards explaining his gruesome platoon splits against them.
I’m still interested in focusing on pitchers who can split the handedness battle in two and develop different arsenals for each, but I think it’s wise to be respectful of how important a good development and coaching staff is for such a tricky approach. Anderson prospered with the Dodgers and has so far floundered with the Angels. Assuming I’m smarter than the Dodgers, that I saw something unique that they missed, is obviously silly. Betting on consistency from a pitcher who had previously made a career out of inconsistency is also a head scratcher. I’m mostly happy with the idea of betting on someone who made huge changes, but I probably should have spent more time thinking about the risks that come with those changes.
I’ve been attempting to identify sleeper hitters for a few years as part of our prospect coverage, but as I mentioned in that writeup this year, the process is getting harder over time. This year, the wheels really fell off. I highlighted Angel Martinez, Osleivis Basabe, Gabriel Martinez, Esteury Ruiz, Yeiner Fernandez, and Jhonkensy Noel. Of that group, only Fernandez had what I’d consider a successful offensive season, though Basabe at least reached the majors on the earlier side of his timeline to salvage a little bit of my dignity. This isn’t what you’d like to see for a list of wRC+’s of hitters I identified as interesting:
I’d like to tell you that this is just an example of good process and bad outcome, but I don’t think it is. The big shortcoming of the methodology I’m using — basically combing through earlier minor league data for skillsets most correlated with eventual major league success — is that it doesn’t incorporate new data. It’s not exactly easy to find, but you can get Statcast-style data for plenty of minor leaguers at this point, and it adds a useful new dimension of analysis.
Ten years ago, the subset of guys that didn’t look like traditional power hitters, posted solid minor league numbers, and weren’t top 100 prospects contained a mixture of true slap hitters and little guys who crushed the ball or barreled it up with great frequency. These days, everyone has gotten better at identifying that second group because we have better data. That makes for adverse selection; the guys with good minor league numbers and good batted ball data are just capital-D dudes. They aren’t flying under the radar anymore. Noel was the only hitter on my list who doesn’t fit into that “weak batted ball category,” but his risks are more of the “yeah but can he hit” variety anyway.
Put simply, I don’t think that the kind of sorting I’m doing is sufficient to identify unnoticed prospects with high upsides anymore, because the kind of guys that this might have captured 15 years ago are now getting noticed. Evaluation has just gotten better, and if I want to find good hitters that traditional evaluators are missing, I’ll have to get better, too. I’m not sure there’s an easy “fix” to the methodology, but I’m excited to try some new ideas this offseason. Which ones? I couldn’t possibly say until I’ve spent more time trying. But this is the preseason prediction I’m least happy with, and I hope to have something better in its place sooner rather than later.
Playoff Teams and Awards
I’ve already talked a decent amount about my predictions because of one oddball: I picked the Orioles to win the AL East in our staff predictions. That one worked out pretty well, I’d say. The total predictions were as follows:
Preseason Predictions by League
I also predicted a World Series of Dodgers over Astros — boring! I’m going to end up doing pretty well in the AL and pretty poorly in the NL, though I think that a lot of people got bamboozled in the Senior Circuit. Everyone on staff predicted that the Padres would make the playoffs, and almost everyone had the Cardinals too.
I think that the lesson I’m going to take away from these is to stick with my method, which is to mostly pick the most likely teams but also try to post a few undervalued options. I didn’t think the Orioles were the most likely team to win the AL East, but I did think that they stood a decent chance of making the playoffs, and also that no one would pick them. I like that juxtaposition, and I picked the Dodgers to win the NL West for a similar reason, believe it or not. Only three of 27 prognosticators had them winning the division, and I thought it was close to a 50/50 shot.
I’m not really sure how to assess the thought process behind these predictions in retrospect, other than to say that if you want to stand out on exercises like this, it’s a good idea to hunt a few undervalued options. You can’t just go rogue on every single pick, and leaning into the Giants instead of the Diamondbacks as my unlikely team in the NL didn’t pan out, but I like to imagine the exercise as a bracket-picking contest, and it’s hard to win if you just go all chalk. Sprinkling in a few 10% chances that the rest of the pickers are weighing as 0% chances is a fun way to do that.
To be honest, I don’t have a lot to say about my awards picks. I didn’t put as much time into those as I did into the playoff teams. I mostly went chalk with Cy Young and Rookie of the Year options, and then picked MVPs based on which teams I thought would have viable contenders on great teams. If the Astros and Mets had delivered huge seasons, I think Alvarez and Lindor would have been at the heart of them a lot of the time.
Thanks for reading through this largely self-centered exercise. I think there’s something to be learned from looking back on old predictions, to hold myself accountable if nothing else. I knew a lot of the conclusions before writing this article, but I still learned more about what I dislike about my sleeper hitter methodology, as well as what I do like about a few other predictions I made. Hopefully when I write something similar for next year, I’m 5% better at the exercise as a result of doing this today. All in all, I think I’d award myself an attaboy, though with some clear get ‘em next times in the mix.