This week marks the halfway point of the 2024 regular season. We’re out of silly sample size season now, having moved on to wondering which teams will add at the deadline and which will start building for tomorrow. Season-long leaderboards are populated with the best players in the league, just like you’d expert. The White Sox and Rockies are awful; the Yankees and Dodgers are great. Plenty of this season has gone according to plan.
Not everything has, though. The Blue Jays and Cubs didn’t get off to the starts they hoped for. On the other side of the coin, the Phillies and Guardians have both exceeded expectations by a mile. Perhaps not coincidentally, both teams have gotten superb performances from their relief corps all season. It’s largely the usual suspects: Emmanuel Clase is one of the best closers of the decade, while the Phillies had the best bullpen projection in the sport coming into the season. But it’s not exclusively the usual suspects. Case in point, or perhaps I should say Cade in point: Cade Smith.
If you’re not a Guardians fan, you might not know who Cade Smith is, and I can hardly blame you. He made his major league debut this season after a solid 2023 campaign during which he compiled a 4.02 ERA (3.42 FIP) and struck out 35% of opposing batters. He struggled to control his walks and Triple-A hitters touched him up for six homers (20% HR/FB), but all told, it was a good year. He broke camp with the big league team; those same power rankings that liked the Phillies so much had Smith down for 61 innings of work as a middle reliever.
That hasn’t exactly transpired. Instead of providing unexciting volume, Smith has been one of the best relievers in baseball. He’s thrown 36.2 innings and allowed just eight runs to score, good for a 1.96 ERA. That might underestimate him, even. He’s striking out 34.3% of the batters he faces. He has his walks under control. His FIP? A spectacular 1.79. Fancier ERA estimators are just as impressed. Smith has a 2.61 xERA, a 2.37 xFIP, a 2.16 SIERA — you name the jumble of letters that aims to use complicated math to explain performance, and Smith rates well by it.
This sounds for all the world like an out-of-nowhere success story. Smith signed with the Guardians as an undrafted free agent. He came into the year as our 27th-ranked prospect in their system. He mostly escaped notice as a single-inning relief pitcher with middling run prevention numbers. He doesn’t throw 100, and he doesn’t have a breaking ball that will make you question how well you understand the laws of physics. This isn’t what elite reliever prospects look like.
That’s all well and good, but here’s the thing: Smith is an elite reliever, at least right now. He’s putting up some of the best results in the game. Projection systems have been quick to buy in; Smith’s projected ERA the rest of the way is 35th in baseball. This isn’t some flash in the pan; he’s getting outs in sustainable ways. That just leaves two very important questions: Where did all of this come from, and how did he put things together so quickly?
As it turns out, there’s a pretty simple answer to both of those questions. Here’s a list of the most valuable four-seam fastballs thrown by relievers this year, using Statcast pitch classifications:
Most Fastball Value, Relievers, 2024
That sheds some light on things. Smith is just punking people with a plus fastball. He throws it more than two-thirds of the time, as you might expect for such an excellent pitch. He gets swinging strikes on 13% of those fastballs, in the 80th percentile for major leaguers. He also generates popups on a staggering 37.5% of his fly ball contact. That’s 16 fly balls, six of which have been popped up harmlessly to infielders, a truly staggering rate. Weak contact and plenty of swinging strikes? It’s no wonder Smith leans on his heater.
If you’re following along, I assume you’re starting to wonder the same thing that I did: How did a guy with a plus-plus fastball grade out as a 40 FV prospect? But then you might do what I did and look at that fastball more closely. That’s when the cognitive dissonance starts to kick in, because Smith’s heater doesn’t appear particularly interesting at first glance.
In terms of vertical break, it’s middle of the pack; Statcast calculates that it drops a scant 0.1 inches less than the average fastball thrown with similar velocity. Speaking of velocity, Smith is hardly a soft-tosser, but he’s averaged 95.7 mph on the pitch so far this year, which is more above average than it is otherworldly. The pitch doesn’t have crazy spin or wild horizontal movement. PitchingBot’s model gives the pitch a 54, Stuff+ a 114 (50 and 100, respectively, are average), and various other public pitching models agree. By the raw numbers, it’s a good pitch, but not a great one.
That said, I think a few things are going on that stuff models aren’t perfectly capturing. The main thing I like about the pitch is the gap between how it looks like it should act and how it actually behaves. There’s some funk in Smith’s delivery. Despite standing 6-foot-5, he releases the ball lower than league average. Here’s how it looks:
And a second opinion:
The aforementioned funk is immediately evident. He steps far down the mound and releases from a low three-quarters arm slot. Here’s one way of thinking about that: Smith releases the ball eight inches lower than his listed height, while the league average is more like 4.5 inches. That’s what you’d picture from a sinker/slider guy rather than a four-seam spammer. That’s not really the arm angle you expect from a swing-and-miss four-seam pitcher who lives up in the zone. In terms of pitchers with mismatched release points and pitch movement, there’s an easy comparison: Smith’s release height is exactly identical to Josh Hader’s.
That’s not a completely idle comparison. Smith’s fastball doesn’t explode like Hader’s does, but it’s a similar beast. He imparts more vertical break than you’d expect from the angle and the speed. Hitters see his release point and expect a two-plane fastball — think Zach Eflin, Kirby Yates, or maybe Grayson Rodriguez on the explosive side of things. Instead, Smith’s fastball is mostly north/south; it tails to his arm side slightly, but mostly it just resists gravity.
That mismatch between expected and actual movement has a lot to do with his fastball’s success, but that’s certainly not all there is. We’ve already noted Smith’s above-average velocity. He gets elite extension, which means that velocity plays up even a little bit more. Put it all together, and we’re talking about a strange pitch. It’s not just me saying that; in our prospect writeup, Eric Longenhagen specifically mentioned Smith’s low-slot, deceptive delivery as a reason that his fastball plays up.
Why don’t pitching models like it quite as much as this piecewise description? I’m not sure, which is part of what I don’t like about pitch models. Luckily, though, Eli Ben-Porat has been digging into pitch modeling in a way that prioritizes explanation, focusing on four factors: unexpected vertical break, unexpected horizontal break, velocity, and extension.
I had a lot of fun working through Eli’s modeling, and if you have the time to do so, I’d suggest following along at home. If you don’t have the time or inclination, though, I’ll break it down for you: Smith’s fastball grades out as one of the best ones in baseball in Eli’s model, and it’s largely thanks to the unexpected vertical break.
More specifically, an average fastball thrown from Smith’s release point creates around 15 inches of vertical break. After adjusting for time of flight, Smith’s fastball rises about two inches more than that. That’s a big deal, worth about 0.4 runs per 100 pitches. For comparison’s sake, that’s similar to Mason Miller’s fastball purely in terms of the value of the break. Miller adds to his with overwhelming velo, of course, but in terms of shape, the two are both great. Another comp? It’s similar to the value Spencer Strider gets from his fastball’s vertical movement. In other words, it’s excellent. Smith’s velocity and extension each add 0.2 runs per 100 pitches, give or take, while his horizontal movement docks him slightly. Put it all together and Eli’s SMOKE model thinks that Smith’s fastball is worth 0.77 runs above average per 100 pitches. That would have been good for 12th in the majors for 2023. Yeah, that’ll play.
Of course, that’s just one way of looking at it, and it’s hardly infallible. The model is heavily based on regressions, as it has to be to work in this fashion. As Eli put it, “it’s not a very sophisticated methodology, but when we’re talking about thousands of pitches in a macro context, it works really well.” I completely agree with that sentiment, and I also happen to think it works well in Smith’s particular case.
The best part of this explanation, to me, is that the things the SMOKE model likes about Smith’s fastball are also the things I like, and they correlate well with the way he’s getting his results. What’s one of his best qualities? All those easy popups. How do hitters produce popups? By swinging under the path of the ball, which happens much more often when they’re expecting a fastball with more east/west movement than they actually get. The extension and velocity are just icing on the cake, putting hitters in a bind faster than they expect. The ball gets on hitters too fast, from too close. It’s moving all wrong. No wonder the combination works so well.
At the end of the day, a model grade isn’t what makes a pitcher dominant or unreliable. The game isn’t played on paper, and plenty of pitches don’t work the way you’d think. Models do a good job in the aggregate, but they’re never the last word. I’m interested in them because I like to use as much data as I can get, but it’s dangerous to rely too much on any one measure, whether it’s computationally clever or simply the eye test.
All that said, it’s certainly not a bad thing that Smith’s fastball is both generally beloved by pitch models and especially beloved by a pitch model that breaks things down into components. It’s even better news that the components single out the thing that I already liked about that fastball: its unique shape relative to opposing expectations.
None of this means that Smith is going to be this good going forward. As great as his primary pitch is, no one can live on fastballs alone. Give hitters enough time, and they’ll adapt to even the toughest offerings. There’s also the question of command; Smith’s been a double-digit walk rate guy throughout his time in the minors, and 35 innings of major league competence isn’t enough to say all those free passes are behind him. At some point, he’ll have to either locate the fastball consistently or start using secondaries more often to keep hitters on their back foot.
Whether or not he’s capable of doing that, though, I think there’s something to learn from Smith’s sudden emergence. When I’m trying to understand what’s driving an unexpected performance, I prefer to come up with a hypothesis first. In this case, I thought that Smith’s fastball shape was leading to his excellent results. I feel comfortable that a closer look at the data supports that conclusion. Cade Smith is great because he has a unique pitch and knows how to use it. Sometimes, it’s as simple as that.
