After a long, cold winter, it’s a welcome sight to have teams back on the field this spring. The game’s biggest stars are getting back into their rhythms, some heated position battles are underway, and — most importantly, of course — new data is pouring into sites like FanGraphs. Caution is always advised when evaluating players based on their spring training statistics, but we can still learn a lot from what happens in these exhibition games.

One of the safest places to start is with a metric that players (pitchers, in this case) have the most control over: velocity. Most statistics are contingent on circumstance, making them less reliable, particularly this time of year. Just about everything a hitter does is a reaction to the pitch coming his way, and most pitching statistics are impacted by the fielders, the umpire, and even chance. How hard a pitcher can throw, on the other hand, is how hard he can throw. It may change as a result of health, age, conditioning, or mechanical adjustments, among other factors, but for the most part, it isn’t dependent on the hitter at the plate or the players in the field. It’s as raw a metric as we have.

That said, I’d be hesitant to read much into which pitchers are throwing softer in the first two weeks in spring training. To some extent, that’s what these games are here for, to build strength and get back in shape, and peak league-wide fastball velocity doesn’t usually come until the warmth of late spring. But the guys who are throwing significantly harder than last year? The ones who are throwing harder than they ever have? That seems worth noting.

Here are the pitchers who have gained the most velocity on their four-seamers so far this spring.

Other than a few high-velocity relievers — Ferrer, Alcala, and Mears — and two swingmen, Mitch White and Joey Wentz, this list consists of starters. Smith-Shawver, Manning, and Gore are former top prospects looking to have breakout seasons. Ober has established himself as a reliable member of an excellent Twins staff, even as he has placed below the 20th percentile in fastball velocity. Meanwhile, Flaherty and Manoah are former Cy Young Award vote-getters whose careers have stalled after injuries and poor performance. The return of their four-seam velocity could be key to their redemption campaigns.

Then there’s Tarik Skubal, whose inclusion on this list should send shivers down hitters’ spines. Skubal missed the first three months of last season after having flexor tendon surgery, but upon his return he was one of the best pitchers in baseball. Last season, his fastball velocity jumped to 95.8 mph, up 1.7 mph from the year before, and the pitch was worth 2.3 runs per 100 pitches, up from -0.3 in 2022. Skubal’s changeup also improved last season, so he increased his usage of both pitches to play them off one another. That combination could be more dominant this year with greater fastball velocity. Skubal has thrown 28 heaters this spring, averaging 97.5 mph and topping out at 99.6. We should reiterate that this is an incredibly small sample size, but considering how much better he was last year with increased velocity, this additional surge could make him even better.

What could this added velocity mean for these pitchers? Well, Skubal’s 2023 is a good best-case scenario, but he was hardly the only hurler who has benefited from a little extra juice. Pablo López’s four-seamer and sinker hung around 93-94 mph from 2019 to 2022 before jumping to 94.9 and 94.5, respectively, last year. Both became more effective, with the four-seamer getting a lot more swings-and-misses and the sinker inducing much more weak contact. A few years back, Carlos Rodón jumped from hovering around 92–93 mph to above 95; his whiff rates went way up and his xwOBAs way down, and he earned himself two trips to the All-Star Game. Alex Cobb’s nice resurgence has corresponded with his unlocking another 2 mph on his sinker when he joined the Giants in 2022; the same is true for Jordan Montgomery.

Naturally, a lot of this is pretty intuitive. Of course adding velocity improved a pitcher’s performance! But it doesn’t always work out that way. For example, Reid Detmers tacked on 1.1 mph to his four-seamer from 2022 to 2023 and fell from the 71st percentile in fastball run value to the 11th. Some players figure out how to unlock the extra velocity but don’t get the results that others do. There are so many variables at play; it’s hardly as simple as more fastball velocity equals better production.

So before we go celebrating this spring’s velocity climbers, let’s take a look at some recent historical data for patterns. Might one type of pitcher stand to benefit more from a bump in fastball velocity than another? In particular, would the softer throwers such as Ober improve more or less from an extra mph of fastball velo than Smith-Shawver, who already throws hard?

Let’s start by asking how much harder the average fastball is to hit at different velocities. We know faster fastballs are more difficult to hit, and some great work has been done on what a pitcher can get out of an extra bit of velocity. But does a one-mph increase at lower velocities have the same magnitude of impact as it does at higher velocities? The data suggest maybe not: Since 2015, hitters have performed worse by greater margins as velocities increase. The difference in performance between four-seam fastballs in the 91–92 mph range and those in the 92–93 mph range has been a mere nine points of xwOBA. Meanwhile, the difference between four-seamers thrown at 96-97 mph and those that are 97-98 mph is 17 points. In this chart below, the bars aren’t declining at an even rate; the slope is getting steeper.

So does that same principle hold when a pitcher’s fastball velocity increases or decreases? To try to tackle that question, let’s take a look at the 430 pitchers who threw 50 or more fastballs in each of the last two seasons and compare how their four-seam fastballs changed (led by Cole Ragans, whose fastball leapt an incredible 4.4 mph from 2022 to 2023) with how they performed. We probably wouldn’t expect these variables to have too tight a relationship given just how much goes into a pitch having success — we aren’t factoring in spin or stuff models or command. But in any case, let’s take a look, with the velocity change on the x axis and the change in xwOBA on the y:

Well, we’d have been right. If you squint, you might be able to see a downward trend in xwOBA as velocity increases, but the truth is, this relationship is pretty darn weak (an r-squared of .008, for those scoring at home). This tells us that any increase or decrease in xwOBA isn’t very explainable by a change in velocity in this data set.

But we have an idea from above that the difference between, say, 91 and 92 might not mean all that much. So what if we cut out the lower-end velocity pitchers and just looked at the harder throwers? Maybe those who reached 94 mph in 2023?

For comparison’s sake, here’s a chart with just those below 94 mph in 2023:

There looks to be a more meaningful pattern for the harder throwers, and the math bears this out: The scatter plot for 94 mph and over has a .045 r-squared, compared to a .001 for the cohort below 94. Now, .045 is still pretty low — again, there’s still a lot we aren’t accounting for — but it’s quite a bit higher than .001. The change in velocity seems to be more impactful at those higher velocities.

The more testing, the better, though, right? I’m not satisfied looking at just one year-to-year change. I’ll spare you the charts, but using a sample of pitchers with the same parameters from 2021 to 2022, the relationship between the change in velocity and the change in xwOBA has an r-squared of .076 — a much higher value to begin with. For those at or above 94 mph in the latter year, that value balloons up to .194; for those below 94 mph, it shrinks to .014. It’s not as if 94 is the magic number, either: for 95+, the r-squared is .287; for 96+, .265; and for 97+, .373.

Without factoring in pitch location or stuff, and with the caveat that this is still a limited sample, these velocity changes have appeared to mean more for harder throwers, at least as far as their four-seamers are concerned. This may be surprising or it may not be. As you get closer to the league’s extreme velocities, it makes some sense that each mile per hour could mean a greater increase in dominance. There’s still a ton of room for any player to buck this trend — a pitcher is more than his fastball, and a fastball is more than its raw velocity — but as we pick apart this spring’s elevated velocities, we may want to lend more credence to the top of the crop.