Rich Hill Has a Theory About Spin and Aging

PHOENIX, Ariz. – You are probably aware of Rich Hill’s story.

You are probably aware of his remarkable comeback and the excellent second act of his career. You probably know he is something of a sabermetric darling, having challenged conventional wisdom with his pitch usage. The A’s and Dodgers were willing to invest in his small sample of success in 2015 and 2016 thanks, in part, to spin rate, which has become newly measurable. Hill also appealed to traditional scouting eyes due to the deception of his delivery and his ability to make his curveball look like two or three pitches by mixing speeds, arm slots, and shapes.

This author certainly finds Hill to be of great interest. So when I visited Dodgers spring-training camp earlier this month, Hill was one of the players I was hoping to interview.

One of the great benefits of being in a clubhouse as a reporter is having an idea arise from the player’s perspective in conversation, an idea from the player’s experience that one had not considered. I presume that is, in part, why teams are more interested in creating conduits between players and the front office, more interested in embedding quantitative analysts in clubhouse.

As I spoke with Hill about where the game is today, he brought up a theory that I found interesting: Hill believes spin ages significantly better than velocity.

“Take for example, David Wells,” Hill said. “Call him up today, I guarantee he could spin a breaking ball pretty well. While velocity may decline, spin I don’t believe declines as fast. The more you have that feel for spin, the longer you can pitch. [A fastball] at 87-88 [mph] with that late life, the perceptual velocity, that spin, will play much longer than a 95 mph fastball without spin or any of that perceptual velocity.”

We can understand why Hill would believe this. He’s had great success late in his career with below-average velocity and above-average spin — and he’s maintained his spin rates in the Statcast era even as he advances in age.

Consider Hill’s average four-seam fastball RPMs over the last three seasons:

2015 – 2,316
2016 – 2,458
2017 – 2,470

Hill’s curveball RPMs have jumped around a bit — 2,682 (2015), 2837 (2016), 2798 (2017) — but have held mostly steady.

If Hill’s theory were true, that spin ages better with age, it could have major ramifications on player acquisition and investment. Driveline Baseball has done significant research on spin, finding that high-spin arms are probably not more likely to be injured than any other type of arm.

Curious about Hill’s point, I enlisted the help of aging-curve guru Jeff Zimmerman to find how spin ages compared to velocity when evaluating two- and four-seam fastballs.

Here’s how four-seamer velocity and spin age, according to Jeff’s data dive. (Note that the following charts depict year-to-year change):

How Four-Seam Fastballs Age
Age Velocity (mph) Spin (rpm)
21 0.0 0.0
22 -0.1 -0.1
23 -0.4 -0.5
24 -0.6 -1.2
25 -1.0 -2.1
26 -1.2 -3.3
27 -1.3 -4.7
28 -1.4 -6.1
29 -1.5 -7.5
30 -1.6 -9.2
31 -1.5 -10.7
32 -1.3 -12.0
33 -1.6 -13.6
34 -1.6 -15.2
35 -1.7 -17.0
36 -1.8 -18.8

And two-seamers:

How Two-Seam Fastballs Age
Age Velocity Spin
22 0.0 -11.7
23 -0.3 0.0
24 -0.7 -25.1
25 -1.2 -40.0
26 -1.1 -58.5
27 -1.1 -59.5
28 -1.0 -69.0
29 -1.0 -73.9
30 -1.1 -67.4
31 -1.3 -85.8
32 -1.6 -92.9
33 -1.5 -105.9

Is Hill on to something? Yes and no.

It’s not true that spin is immune from the effects of aging curves. Pitchers generally lose spin as they age. But the four-seam fastball loses a little less spin than it does velocity, remember the average fastball is about 93 mph with 2,200 rpms. The two-seam fastball loses more spin than the four-seamer, but since sink is tied to lower spin, that could be a good thing.

Hill is correct in that spin rate has become an important evaluation tool, even below the professional level.

While velocity has long been the key measurement to begin the process of weeding out pitching prospects since the advent of the radar gun, now teams have this other tangible measurement.

Amateur prospects are now monitored with TrackMan radar at showcases. A subset of teams appear to be making decisions informed by this data. And if teams find that a high-spin, lower-velo pitcher is a safer bet to age well, to be healthier, there is perhaps significant value that can be unearthed in valuing spin over velocity. That’s Hill’s theory.

“You are going to see guys that get drafted in the first round throwing 88 mph,” Hill said. “There’s something there that is much more valuable than a 95-96 mph fastball. [Spin] probably projects to play longer… I mean the ability to stay healthy, the ability to play for a long time. [With spin,] you are investing in something for the long haul as opposed to just a few years.

“Now, if you have a guy who throws 95-96 [mph] who has carry on a baseball, that’s really special.”

We all have biases tied to our experience. Hill’s experience is as a pitcher who has had an amazing second act and who has not lost any spin since Statcast’s doppler radar has been watching.

However, it’s not true that spin doesn’t decline with age. Fastball velocity and spin both immediately decline for most pitchers. Few things age well, after all. But if there’s something to staying healthier by focusing on feel for spin instead of maxing out velocity, maybe Hill is on to something. The last decade is littered with high-velocity arms that have undergone Tommy John surgery. With only three years of Statcast data, we’re going to need more information.

But even if Hill’s experience is an outlier, it’s still worth exploring.





A Cleveland native, FanGraphs writer Travis Sawchik is the author of the New York Times bestselling book, Big Data Baseball. He also contributes to The Athletic Cleveland, and has written for the Pittsburgh Tribune-Review, among other outlets. Follow him on Twitter @Travis_Sawchik.

29 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments
Captain Tenneal
5 years ago

So 4 seam spin doesn’t decline at all? 12 RPM over 12 years is literally nothing

Pwn Shop
5 years ago

Not to mention this begs the question, how about curveballs, i.e. Hill’s bread-and-butter.

johansantana17
5 years ago
Reply to  Travis Sawchik

So overall spin rate decrease from age 21 to age 36 is the sum of all the numbers in the chart (-122 rpm), rather than -18.8 rpm?

Is that true for velocity too? If so, that would mean that fastball velocity drops 18.6 mph from age 21 to age 36, which seems unlikely to be true.

johansantana17
5 years ago
Reply to  Travis Sawchik

I don’t really understand. What type of effect would that have on the data?

Joe Joemember
5 years ago
Reply to  johansantana17

Data appears to have survivor bias, but still gets the point across. Players that have velocity drop at earlier ages don’t remain in the league. Players that have a huge drop in velocity in a year affects that year, but likely wash out so they don’t affect next year. In other words, adding up the velocity drops isn’t applicable as guys still in league in their 30s tend to be the guys that don’t have typical velocity drops.

Ryan DCmember
5 years ago
Reply to  Joe Joe

Survivorship bias doesn’t explain the data, if Travis is correct that this is year to year rather than cumulative comparison against a baseline. It beggars belief that the average 37 year old throws 18.6 mph slower than the average 21 year old. If anything, survivorship bias would cut the other way, because only the 37 year olds with the highest velocity (relative to other 37 year olds) would be viable major leaguers. Rich Hill throws in the upper 80s, not the upper 70s.

Joe Joemember
5 years ago
Reply to  Ryan DC

9 players throw 90 mph at at age 21.
3 of them throw 88 at 22 and are cut.
3 of them throw 89.
3 throw 90.
Average drop in velocity is 1 mph from 21 to 22.
6 remaining 22 year olds have average mph of 89.5.
2 guys throw 87.5 at 23 and are cut.
2 guys throw 88.5 and 2 throw 89.5.
Average drop is 1 mph
4 remaining 23 year olds that pitch as 24 year olds had average mph of 89 at 23 instead of 88.

Edit: Math may be off. Beer and math are not friends.

Justin Cmember
5 years ago

Yeah, the axis scales are way off. If the average fastball is 93 mph and 2200 rpm, a 1 mph decline is -1.1% and a 10 rpm decline is -0.5%.

The lines really diverge even more than they’re showing. A 1 mph decline is equivalent to a 24 rpm decline in spin.

johansantana17
5 years ago

Yeah, the differences in scale on the first graph are really deceptive. They make it seem that the magnitude of the decrease is similar, but -13 or -14 rpm of spin rate (where the average is 2200 rpm) is a significantly smaller decrease than -1.5 mph of velocity (where the average is 93 mph).

13.5/2200 = .00614

1.5/93 = .0161

.0161/.00614

Velocity declines at a 2.6 times quicker pace than spin rate.

synco
5 years ago
Reply to  johansantana17

Yeah I mean from the numbers it looks like Hill is resoundingly correct. Velocity declines much more quickly than spin rate, which is exactly what he’s quoted as saying.

channelclemente
5 years ago

Particularly when you realize a FB spinning at 2300 rpms makes around 10 rotations in it’s transit to the plate.