The Velocity Surge Has Plateaued
Among the proposals exchanged by MLB and the MLBPA was the idea of studying the mound. More specifically, baseball is interested in studying what might happen were the mound to be lowered, or were the mound to even be moved back. In a sense, adjusting the mound might seem radical, but of course, the mound has been lowered before, and baseball wants to see if it might be able to combat the ever-increasing strikeout rates. The league-average strikeout rate in 1998 was 16.9%. A decade later, it was 17.5%. Yet a decade later than that, it was 22.3%. That’s a 27-percent increase in strikeouts over the course of ten years. You can see why people might want to nip this in the bud.
Why have strikeouts been on the rise? How might you explain all the swinging and missing? I suppose there are the people who might just grumble the term “launch angle” and leave it at that, but a more compelling explanation might be the league-wide increase in velocity. It’s been hard not to notice — as they say, now every bullpen has a half-dozen guys who come out throwing 95 miles per hour. Billy Wagner used to throw 96. Now everybody throws 96. And the less time hitters have to react, the more often they’re going to whiff. Case closed! We’ve all solved it, together.
Except for the part where velocity has ceased increasing. The velocity surge was something I think a lot of us just took for granted. Teams like velocity, and players are training harder than ever before. But the surge has slowed, if not stopped. You don’t need to dig too deep to find evidence.
What follows is information from a few different sources, but I’ve done what I could to ensure across-the-board consistency. Let’s start in the obvious place — league-average fastball velocities, from 2008 through 2018, courtesy of Baseball Savant:
Even the surge itself is maybe not that eye-popping, when you look at it like this; from 2008 to 2016, the average fastball got faster by about a tick and a half. We’re not talking about everyone going from 85 to 95 overnight. Still, consider that the major leagues are selective for the most talented arms in the world. These are pitchers several standard deviations above the greater population mean. For upper-level fastballs to get faster by a tick and a half is pretty remarkable. But, since 2016: nothing. 92.9, 92.8, 92.8. The surge has flattened out.
We can break this down in several other ways. Here’s a plot showing the league rate of fastballs thrown at least 95 miles per hour:
There’s a very steep increase, before the line levels off. You don’t see much movement between 2008 – 2011. Then, from 2012 – 2015, there’s an increase of more than eight percentage points. Then the recent flat bit. In 2015, 21.7% of fastballs were thrown 95+. In 2018, that rate held at 22.1%. Nothing doing over four years. It is, of course, relevant that there was also nothing doing over a four-year span a while ago, right before the real surge, but that doesn’t mean another surge is coming.
Related to the above, we can look at the following plot, showing the rate of pitchers averaging at least 95 miles per hour with their fastballs:
There’s a clear overall increase, even despite a few year-to-year drops. And that’s what we’re looking at in 2017 – 2018: a year-to-year drop. But, from 2015 – 2018, there’s an increase of just 1.5 percentage points. From 2012 – 2015, there was an increase of 4.8 percentage points. From 2009 – 2012, there was an increase of 4.5 percentage points. It can feel like everybody on every pitching staff throws 95 miles per hour. In reality, at least in terms of average fastballs, only about a fifth of pitchers qualify. That fraction has hardly budged of late.
Switching to bars for a moment, let’s take a look at raw counting stats, for elite-level velocity. Here’s the league-wide number of fastballs thrown at least 100 miles per hour:
Last year had a greater number than the previous year, but it also had a lower number than the previous year. There hasn’t been much overall movement since 2015. Rare are the pitchers who can reach triple digits, and rarer still are the pitchers who can stay there without getting hurt. Some bodies just can’t train themselves to 100, even if the pitcher does everything right. We’re pushing up against certain physiological limitations.
For one last plot, here are the league-average fastball velocities for rookies, and rookies alone:
As the league increasingly looks for velocity, you’d expect lower-velocity pitchers to be replaced by higher-velocity pitchers. You’d think that, in theory, baseball would be selecting for pitchers throwing harder and harder. But that’s not what the plot is showing. From 2008 – 2013, rookie fastball velocity increased by a full two ticks. And for six years, it’s held in the same place. Compared to 2013, 2018’s rookie velocity was up 0.18 miles per hour. That’s not literally *nothing,* but it’s nothing to write home about. Rookies aren’t all emerging from the minors throwing gas. Not, at least, at increasing rates.
The easy thing to do is to observe what’s already happened. In the past, we observed that there had been a velocity surge. Now, we observe that the surge has flattened out. It’s happened somewhat quietly, but I don’t know another way to interpret the data above. The hard thing to do is to forecast the future. Have we reached only a false summit, with another climb just in front of us? Or is this just going to be it? On the one hand, pitchers continue to train, and they continue to train hard. Teams everywhere are investing in tech, trying to find the keys to efficient biomechanics. Facilities like Driveline are becoming increasingly popular. On the other hand, a given human body can throw a baseball only so fast. Beyond a certain threshold, bodies stop working. We might well be approaching a point where major-league pitchers are throwing at almost 100% efficiency. And as much as teams might want to select for great velocity, great velocity doesn’t make one a great pitcher by itself. There have to be more considerations.
Every team could have a harder average fastball, if that was all they cared about. Every team has a bunch of hard-throwing young arms in the minors, and the reason they’re not in the majors is because they’re not ready. If there is to be another league-wide velocity increase, perhaps it might happen because teams figure out how to develop good mechanics more effectively. Maybe that’ll come out of all their new tech. It’s not yet something we can comfortably take for granted. For now, we can say that big-league velocity isn’t out of control. It seemed like it might be for a number of years, but pitchers can do only so much before they’re all doing as much as they can.
Jeff made Lookout Landing a thing, but he does not still write there about the Mariners. He does write here, sometimes about the Mariners, but usually not.
I really hope a study of the mound takes place in the near future. Seriously, who doesn’t like studies?
“who doesn’t like studies?”
I am not pleased to point this out, but I imagine that might be answered with demographics; among the sabremetric-minded community, perhaps relatively few, but among the general population, I might guess as much as 50% dislike ‘studies’, as that is approximately the portion who affiliate with organized religions which espouse beliefs and worldviews that are thoroughly incompatible with science, logic, rationality, and intelligence.
Wat
Although you maybe right, it’s a rather “heavy” response to a not so serious question
“Who does not like to learn?”
I’m glad you reported it, but I think we can respond to demographic data; Among the oriented communities, it can be relatively small, but in the general population, it seems that 50% do not like the “search”, and the behavior that relates to the religious organization that I recognize. Vera and Quiet are muscles that are not compatible with science, logic, reason, and intelligence.
I think we should really be talking about how that’s an incredibly long run-on sentence.
You trying to say that Jesus Christ can’t hit a curveball?
“You trying to say that Jesus Christ canāt hit a curveball?” LOL!!!!
u wot m8
The NFL doesn’t like concussion studies.
Perhaps someone should do a study of how many people like studies.
Are you sure that hasn’t been done already?
I’m really curious to see this analysis on secondary pitches. FB velocities have plateaued, but if slider velocities are still trending up, that would seem to predict a continued rise in strikeouts.
Indeed.
A breakdown by pitch type might shed some light.
(Pitch mixes have changed.)
Also a breakdown by starters vs relievers.
(More innings going to all-out relievers, less to starters who might be pacing themselves less.)
One possibility is that training regimes have changed over the past few decades and that the increase might reflect a generational shift as the older players, trained differently, cycled out of the system, resulting in a new baseline.
I do think that’s a very pertinent question. Not so much the velocity range and peak on its own, but more the correlation between the strikeout percentage and the velocity in those two groups.*
We (I) make the lazy assumption that relievers can operate at max velocity and consequently their K-worthy pitches get an appropriate number of Ks. This could be so. I don’t know. But I’m unaware of any study that lends credence to the idea, across the two spectrums of starting pitchers and relieving pitchers.
* You could extend this question to other things. Velo differential between primary and secondary pitches, for example. Spin rate. Even in-zone and out-of-zone strikeouts. Maybe some of this has been covered, but I think there’s a lot to be explored here.
I think this is part of the issue. Pitchers are learning how to throw faster pitches with more movement. Without major advances velocity will essentially cap out and movement will be where pitchers need to go to get greater advantages.
One question – I’d be fascinated to see if the standard deviation of pitch speeds (including FB’s) have changed over time, both for given pitchers, pitchers in an average game, and baseball-wide.
If hitting is timing, and pitching is messing with hitter’s timing, than it would suggest that even a meaningful population subset throwing at significantly higher speeds would raise the overall variance of pitch speeds; as a hitter, you thus have to account for a higher potential range of pitches when you swing, which might help explain the higher K-rate.
This is pure speculation!
But, a very good thought.
did you control for starters vs relievers? cause relievers are obviously making up an ever larger share of pitches in the league.
Average starter fastball velocity has barely moved in three years. Average reliever fastball velocity has barely moved in four.
What about during the “boom” years? Were they in lockstep too?
I wonder about the top 20% RP only? Clearly teams are using bad RPs to mop up more than ever, which should depress avg velo.
Even if the velocities have plateaued, that doesn’t mean that it isn’t already a problem. Let’s assume that this is the new normal for fastball velocities. Let’s assume that the strikeout rates are also at the new normal. (Those are two big assumptions on my part).
The strikeout rate as Jeff mentioned has already increased by a ton. It still behooves MLB to look at ways to reduce the strikeout rate. If lowering the mound or moving it back slightly reduces strikeout rates, then that’s something worth considering. Baseball is more interesting when the ball is put in play.
Deaden the ball too!
I would rather watch a nasty fastball for a strikeout than a routine grounder to short every time.
I went researching for info on the fastest pitchers of bygone eras and came up pretty empty. Everybody here has probably seen the rather poor attempts to measure Feller and Ryan, none of which appear to generate much confidence in their accuracy, but are there any studies, or even educated guesses, about what the fastest pitchers topped out in the distant past, Ed Walsh, Walter Johnson, Satchel Paige, Sandy Koufax and Bob Gibson just to name a few.
Walter Johnson there is some video that has allowed for estimates that suggest he crowded 100. More, he had a weird mostly sidearm delivery. A total outlier for any era which is why he “got by” with essentially one pitch for most of his career.
Gibson there is a lot more modern video. He was plenty fast. But he was also just plain mean. Made him that much more effective. Nobody could get comfy around the plate.
The issue, though, is league average velocity, not just the outliers at the fringes.
I am equally interested in the same thing. What was the average fastball in each era? Did Johnson really throw 10+ mph faster than anybody else and was that also true about Feller? I would love to find out. Since there is very little information even about the greats, there is next to nothing about the solid starters. I grew up locally with Newcombe, Raschi, Reynolds and Radatz and while these men were tremendous pitchers I have no idea how hot their heaters were compared to the other stars of their day like Lemon, Wynn, Bunning, et al.
If 3 batter rule kicks in avg velo goes down
Doubtful. Having relievers pitch to more than 1 batter will allow teams to have relievers pitch more innings.
Eh. Specialist relievers are mostly about arm slot and pitches disproportionately effective against same handed batters. I don’t think velocity is a differentiating factor.
Indeed. If anything, velocity will go up, because that mostly plays against both sides.
That rule can never happen. Its absurd.
I think the rise in velociraptors is now evolving into pitching smarter: breaking ball variances; spin rates; location and command; release points; pitch tunneling is a thing; sequencing; platoons; and valuing catcher game calling, framing, blocking, etc…
The rate of technical innovation is so rapid now and front offices and management are like silicon valley.
Pitchers hunt in packs.
Clever girl
I feel like the game has fewer pitchers who work as āsoft tossersā than 10-15 years ago. I think of some of the best pitchers in the game were effective at working the zone and changing speeds so I think that also skews the data. They were also logging a ton of innings. Maddux, Glavine, Hudson, Zito, Moyer, Wakefield, Pettitte, Buehrle, Mussina, Lowe, etc.
Yes a bunch of guys are throwing 95+ but fewer innings are being thrown by guys with FBs <90.
That’s kind of implied with the rise in K% from 10-15 years ago
When the mound was lower before the1968 season, offensive numbers from the 40s-60s were astronomical. That how we got all these unbreakable records:
Wilson 191RBI
DiMaggio 56 straight
Incredible BAs, SLG, OPS, and all the advanced stats that people have gone back and figured out.
It was like an arcade game. In ’68 they raised the mound too much, obviously, with all the crazy pitching lines that came about. They fixed that, and here we are.
The mound height has nothing to do with the K rates in my opinion. Every dang fool down to the 5’6″ 160lb shortstops are swinging from their shoetops nowadays. It’s that famous line from the PED Era – CHICKS DIG THE LONG BALL.
Guys have never learned to hit properly at any level anymore – just bombs away. Every MLer can hit the fastest fastball, but they’ve never learned how to handle off speed and breaking stuff anymore. Nobody inside outs balls the other way or tries to hit behind runners, and stuff like that anymore. Or learned to hit intelligently ala Tony Gwynn or Wade Boggs.
The freaking increase in punchouts is coming from offspeed and breaking stuff. Mound height has nothing to with that, it only makes all the fastballs more or less effective depending on raising or lowering the mound. That’s not gonna teach dudes how to hit offspeed or breaking stuff unless they are taught young, and reinforced as they get older. Everybody’s bat control stinks because of all this.
tl:dr Until batters are taught to hit and not just sit on the fastball, strikeouts are just gonna keep going up.
Just my two cents. ?
I’ll tell the guys getting paid minimum 500k to hit baseballs that they’re doing it all wrong.
That is exactly the line of thinking that can lead to worsening of problems.
This isn’t even factually correct; the mound was *lowered* following the 1968 season, not raised. And this happened because the 1960s was an extremely pitcher friendly era, not the era that followed.
Hang on… did you just invent history? They didn’t raise the mound before the 1968 season.
Setting aside some of the historical inaccuracies, the game has evolved. Hitters now could go back to the way they were hitting in the 1960’s and 70’s and cut down on their swings and try to punch the ball through the other side. They could bunt guys over and hit-and-run more. They could steal more bases while also giving up a lot more outs (go check out stolen base success rates in the 60’s and 70’s). Players today would strike out less. . . .and teams would score a lot fewer runs. It’s not that strikeouts don’t matter, they do matter. Anyone who has ever said that was just plain misinformed. It’s just that you need to look at the total package of what a player does, not just strikeouts. HR are very, very valuable and offset a lot of strikeouts. Of course, completely selling out for HR isn’t smart either – hitting 35 or 40 HR isn’t too valuable if you are hitting .196 with a .250 OBP.
I do think though that with shifting, we are going to start seeing a move to more balanced hitters who can hit to all fields. It’s just a logical “evolution” in the continual cat-and-mouse of pitching/defense and hitting.
Gallo hit .206/.312 with 40 HR and was worth 2.8 WAR. Just sayin’
WAR is arbitrary – just saying. Put a negative weight on strikeouts and that all changes. Its all a game of approximations. In the end Rangers players, fans and decision makes need to ask themselves how much they value Joey Gallo. if they decide that Joey Gallo is 2.8, then that’s great. Personally I would go with a 1-10 scale or a letter grade, but I am also not constrained by an algorithm.
I don’t think today’s athletes could go back to how it was. Those players had different skills than these do. Joey Gallo is never going to be a station-to-station guy. I actually believe that a lot of good baseball players don’t get opportunities these days. I am not convinced that a quality 80s team would not win over a current team. I would say 70s but I don’t really know that era too well, so I won’t. The term “evolved” implies a process of evolution through natural selection and I am not sure that ever happened – its more like a different evolutionary fork. Are the Cichlids of the fresh water lakes in Africa better at fish stuff than their ancestors? I would say they are just different. As you can tell, I don’t like the term evolve because it carries a lot of bias.
When the hill was lower before the 1968 season, the number of attacks from 40 to 60 – were astronomical. We have all Unbreakable disks:
Vilson 191 RBI
DiMaggio 56 hetero
Incredible bass, SIG, operation and all advanced statistics that people have come back and figured out.
It’s like a video game. They also grow mounds, of course, all the crazy lines have been established. They fix it and we are there.
Hum’s height has nothing to do with K. Foot IMO, any circular passage in 5 minutes, with short stops of 6 160 and 160 lb. Full of beating these days. This is a well-known series of PED-Chicks that digs long balls.
Boys never hit him at a higher level than just bombs. Miller can quickly hit the ball as fast as possible, but I never know how to deal with speed and more. Anyone who throws bullets on the other side, or tries to hit behind the runners, etc.
Fist blows slowly grow and break things. The height of the humm has nothing to do with it, it’s just a matter of more fastball or less efficient depending on the lift or reduction of the hum. Do not learn to get to such speeds or to break things unless they have learned, they are young and grow with time. All blind mice control the smell.
TL: As long as the tests do not learn to hit, and not only fast to sit down on the ball, Promaljaje goes.
Only my two cents. ?
This is a good point, but nobody wants to hear it. K rates are not all because pitchers are exceptional – a lot of it is because hitters are not good at making contact. Probably more accurately, they are not being discouraged from striking out. Also of some accuracy, players with poor hit tool now have a place on MLB rosters, whereas they used to not. But yeah, its all science and training…
Seems to me that the findings in the article are consistent with the idea that while fastball speed for fastball pitchers is rising, there has also been new emphasis on pitchers with good curveballs. And since those pitchers with the good curveballs were selected because of their secondary stuff, they don’t necessarily have great fastballs. This might be causing the decline in the speed of the average fastball league-wide, while not necessarily causing a decline in fastballs by fastball pitchers.
Iām guessing that Aroldis Chapman not living above the 100 mph barrier anymore single handedly changes the raw data on 100mph pitches.
Another contributing factor could be that less fastballs are being thrown, even by the 100 mph club members.
I’m not really convinced that it’s all “training harder.” I think a lot of it is just “throwing with max effort on every pitch.” Everyone knows that velocity gets paid, and I think just about every pitcher from high school on up is throwing as hard as they can on every pitch, Tommy John surgery be damned.
Anyone who has ever pitched competitively knows this isn’t true. You can’t have any amount of control whatsoever if you throw 100% on every single pitch. You have to find the right effort level that balances velocity and command and that fits the situation you’re in (starting, closing, single batter, one inning, etc.)
It’s usually a range of something like 70-90%
Ah, so you play in MLB?
I’m convinced that you are more right than wrong.
There has been a study on lowering mound height. Surprisingly, it found no difference in pitch velocity or movement. The only significant effect was lower stressors on pitcher’s arms from a lower mound. http://bit.ly/2TQfuar
Nothing that can’t be remedied with a change to the methodology.
It’s not that often that I’m surprised by a league-wide trend, but this surprised me.
You can’t just keep throwing harder because you eat more veggies and do more stretching. I would argue that the velo spike has a lot to do with an enlarged strike zone. Guys who previously couldn’t throw enough strikes now can play MLB baseball. An enlarged strike zone in conjunction with shorter usage of everyone has probably gone about as far as it can go. Its fun to promote science and technology and a narrative about bigger, stronger, faster but everything has a point of diminishing return.
Jamie Moyer last pitched in 2012. I’m convinced that alone accounted for the rise we see from then to 2014.