Taj Bradley’s Star Turn

Entering the 2023 season, Taj Bradley was the no. 36 prospect in baseball, a 22-year-old ace who overmatched his opponents to such a degree that he forced himself to the majors for the back half of the year.

In 2025, Taj Bradley was traded straight up for Griffin Jax, a 31-year-old reliever who has accrued exactly zero wins above replacement for the Rays since that trade.

In 2026, Taj Bradley has been one of the best pitchers in the major leagues.

That’s some roller coaster. And while my first instinct is to take Bradley’s first four starts with a giant grain of salt, this isn’t your average “random dude has good stretch” story. Bradley truly is one of the most dynamic pitchers in the world. He’s electric on the mound. He wasn’t a 55-FV prospect by accident. So let’s take a look at what he’s changed, what he hasn’t changed, and whether this recent run of dominance looks like the portent of a new skill level or just a blip on the graph.

As a prospect, Bradley pounded the zone with a mid-90s fastball. He throws from a high arm angle and generates near-perfect backspin as a result. I think that evaluators were probably a bit too high on his fastball as a prospect, because I don’t think we had the same understanding of the importance of pitch plane then. Despite great velocity and movement, his pitch has never been a bat-missing force. He generates a league-average rate of swinging strikes with it; backspin all you want, but Bradley’s high release point means that his pitch doesn’t have the shallow approach angle to miss bats at the top of the zone.

In the minors, this wasn’t evident; Bradley’s velocity and command pretty much knocked the bat out of opposing hitters’ hands. But in the bigs, opposing hitters were up to the task. His fastball has been hit around throughout his major league career, with double-digit homers hit off it in each of his first three seasons. Many pitching prospects encounter that specific obstacle – these major leaguers sure can cover fastballs – but Bradley has been particularly susceptible to it.

A quick side note: Both PitchingBot and Stuff+ rate Bradley’s fastball highly. I was curious why, so I cracked open PitchingBot and shook it around until some insights dropped out. The model doesn’t give Bradley’s fastball high marks when it comes to missing bats; it has a below-average expected swinging strike rate, an accurate reflection of reality. However, it gives Bradley’s fastball high marks in two other fields: drawing called strikes and limiting contact quality. In particular, the model thinks his fastball should induce more popups and grounders, fewer line drives, and less hard contact than your average four-seamer. That hasn’t played out in practice, obviously, but I see where it’s coming from, and I’ll be watching to see if that starts happening in games.

In any case, Bradley’s fastball hasn’t changed much this year, so we’ll have to look elsewhere for changes. First stop: his other best pitch as a prospect, a cutter that he can run all around the strike zone or snap off as a chase pitch. Bradley made an excellent change to the pitch over the winter. He altered its shape significantly, losing ride and gaining glove-side break. He also took a mile an hour off of it, even while his fastball has sped up. That creates a larger movement differential – it separates from the fastball by three inches more than it did last year – and a pitch more suited for attacking the bottom of the strike zone. He’s drawing more chases, and batters are making less contact. Both PitchingBot and Stuff+ love this change, and I completely agree.

Bradley uses that cutter about 20% of the time. He also mixes in a low-90s splitter at the same rate, and the splitter is yet another pitch he’s changed meaningfully this year. This one is simpler. Last year, he had a hard time killing lift on the pitch. He generated 5.5 inches of induced vertical break, too much even for a hard splitter. For reference, Yoshinobu Yamamoto’s similar-speed offering gets about 1.5 inches of IVB. In 2026, Bradley’s splitter has lost nearly two inches of IVB, which means it separates more from the rest of his offerings, back to the shape he threw it with in 2024.

This version of his splitter is turning batters into pretzels despite inconsistent command. For example, this is not where you want to locate a splitter with two strikes:

He threw 25% of those pitches in bad, mid-plate locations, hugely more than league average (10%). But that’s par for the course for Bradley; this is just how he throws his splitter. When he throws it below the zone, though, it vanishes:

And when he leaves it up, he gets his fair share of takes. The pitch below was overturned on review to a strikeout:

When opponents do put a splitter in play, it’s often on the ground. His 56.1% groundball rate on that pitch is the highest in his arsenal, and that informs how he uses it. All told, it’s been his best pitch this year – it’s been great at missing bats, great at drawing bad takes, and also great at inducing weak contact. You can’t really do it much better than that.

For the sake of completeness, I’ll also mention Bradley’s curveball. While everything else in his arsenal is hard – a fastball that touches triple digits, two secondaries that sit around 90 mph – his curve is slow and looping. He uses it frequently as an out pitch, and it helps the other out pitches play up; high cutters and splitters look very similar to low curveballs out of his hand. I’m not enamored with the pitch on its own, and he hasn’t made a ton of changes to it this year, but I do like the way it pairs with the rest of his arsenal.

These pitch modifications are mostly small changes around the edges, but they add up. When every single pitch gets a little sharper, it’s not just a linear effect. In isolation, a small improvement wouldn’t be a huge deal, but our pitching models agree that Bradley’s stuff has improved by roughly one standard deviation in the aggregate, the result of every pitch getting a bit better.

There’s a big mystery here, though. Bradley is missing more bats, no doubt. But he’s still not missing many! Out of 43 pitchers to throw 20 or more innings so far this year, Bradley’s 11.4% swinging strike rate is 22nd, dead average. His called strike rate is 30th. And yet he’s striking out 31.2% of opponents, ninth best out of that group. In other words, he’s running hot on the strikeout front compared to its component statistics.

There’s an easy counterpoint here, though. Bradley has always done this. To use a larger sample, I looked at his whole career. There have been 64 pitchers who’ve thrown 400 or more innings since the start of 2023, his first season. Bradley is 17th in strikeout rate, 25th in swinging strike rate, and 61st in called strike rate. In other words, he’s always done better than you’d expect merely from looking at how many bats he misses and how many hitters he freezes.

How is that possible? I think it has something to do with his north/south attack. He throws from a weird arm angle. His pitches don’t break the way most pitches of a similar type do; his curveball is pure 12/6, his cutter acts like a gyro slider, and his fastball has nearly no arm-side fade at all. The timing’s all messed up. Balls aren’t where opponents expect them to be. Bradley gets a lot of awkward swings, a lot of foul balls, a lot of at-bats where his opponents smack two crushable fastballs foul and then wave meekly at a diving cutter.

Is that going to keep resulting in a 1.25 ERA and 1.87 FIP? Of course not – but I assume you didn’t need me to tell you that. It’s so early in the season that ERA and FIP are still phenomenally noisy. Bradley hasn’t allowed a home run yet, for example, and that’s not going to be the case by season’s end. He has a 90% strand rate – league average is 72%. Luck is going to even out somewhat. I don’t think he’ll keep striking out as many batters as he has so far, even. But a slight regression from this version of Bradley would still leave the Twins with a very good pitcher.

The big takeaway here is that regression looks like it’s going to push Bradley to an ERA with a three handle, a massive improvement on his career mark to date (4.67). He’s not a completely different pitcher. He’s still, in large part, the same guy he was last year: high slot, fastball-first, secondaries that work the strike zone vertically. But when you turn everything up just a little bit – an inch of drop here, an inch of ride there, faster fastballs and slower cutters – the results follow. Bradley has always struck out more batters than you’d expect from his swinging and called strike rates – and now he’s getting more of both.

As an aside, this is a great example of why the whole “never trade with the Rays” thing is overblown. The Rays make a ton of trades. They win more than their fair share, because they’re very good at player evaluation. But talent level isn’t stationary, and no one knows the future. Sometimes players improve unexpectedly. Sometimes players get worse unexpectedly. Sometimes teams have different timeframes and different needs. The Rays needed relief pitching last season, and Bradley wasn’t in good form last summer. It doesn’t have to be more complicated than that.

The bottom line is this: I don’t think that Bradley has completely reinvented himself. He’s not suddenly the best pitcher in baseball. But he’s a great example of the long arc of pitcher development, and a good reminder that small changes can have a big impact. In a lot of ways, Bradley is the same guy he always was. But that guy was already pretty close to stardom. It certainly appears like dialing up all of his pitches by just a notch or two was enough to turn him from a league-average arm into an excellent starter.