How a Baseball Star's Tricky Pitch Strikes Out Hitters—and Baffles Physicists

By Guest Blogger | November 14, 2012 3:07 pm

Andrew Grant is an associate editor at DISCOVER. His latest feature, “William Borucki: Planet Hunter,” appears in the December issue of the magazine.

Last night Major League Baseball announced the winners of the Cy Young Award, given to the year’s best pitchers in the American and National leagues. The National League victor was New York Mets pitcher R.A. Dickey. That he won the award is remarkable, and not just because he is a relatively ancient 38 years old or because he plays for the perennial punch line Mets. Dickey is the first Cy Young winner whose repertoire consists primarily of the knuckleball, a baffling pitch whose intricacies scientists are only now beginning to understand.

Most pitchers, including the other Cy Young finalists, try to overwhelm hitters with a combination of speed and movement. They throw the ball hard—the average major league fastball zooms in at around 91 miles per hour—and generate spin (up to 50 rotations a second) that makes the ball break, or deviate from a straight-line trajectory. Dickey does neither of those things. Rather than cock his arm back and fire, he pushes the ball like a dart so that it floats toward the plate between 55 and 80 mph. The ball barely spins at all—perhaps a quarter- or half-turn before reaching the hitter.

That lack of rotation turns out to be the reason Dickey can get away with throwing the pitch more than 85 percent of the time. A baseball is not perfectly smooth—it has stitches that rise about seven-hundredths of an inch above the skin of the ball. When a typical pitcher throws, the ball spins so quickly that those stitches become irrelevant; it moves through the air as if it were perfectly round. But when the ball barely rotates, that slight protrusion becomes very important. The seams create turbulence as they cut through the air, leading to forces that can push the ball in any direction. “The movement is all over the place,” says Alan Nathan, a physicist at the University of Illinois at Urbana-Champaign who studies the physics of baseball. “It’s the only pitch that’s as likely to go up as down and as likely to go left as right.”

This randomness is the knuckleball’s most endearing yet perplexing trait: Nobody, including the pitcher, knows where and how much it will break. (This is the theme of the critically acclaimed documentary Knuckleball! that premiered in September.) It also leads to a sort of mystique surrounding the pitch, making it difficult to separate reality from perception. Hitters often describe the pitch as dancing or fluttering through the air before making a sharp turn at the last moment. Until recently, there was no evidence to determine whether that is really the case. Fortunately, Dickey happens to be playing in an era in which we can study the physics of his handiwork and measure what the ball really does during its 0.5-second journey between pitcher and hitter.

That’s exactly what Nathan did earlier this year. He tapped into the wealth of data from PITCHf/x, a system of cameras that tracks the speed and trajectory of every pitch thrown in all 30 major league stadiums. (Sportvision, the company that makes the technology, is also responsible for the yellow first down line on football broadcasts and the ill-fated glowing hockey puck.) The cameras capture 60 images a second and can identify the location of the ball with half-inch precision.

After analyzing trajectories from two games pitched by Dickey and two by Tim Wakefield, a Boston Red Sox knuckleballer who retired after the 2011 season, Nathan was able to uphold one piece of conventional wisdom while striking down another. His measurements confirmed that the knuckleball is indeed the most random of pitches—there is no tendency for the ball to break in any particular direction. But his analysis also dispelled the notion that the ball dances to the plate on an uneven trajectory. His measurements revealed that a knuckleball takes just as smooth a path as any other pitch.

Nathan’s findings lead to an interesting question: What accounts for the disconnect between what hitters say they see and what is really happening? Nathan doesn’t have the answer, but he suspects it has more to do with psychology than physics. Hitters aren’t used to seeing the pattern of the seams as the ball approaches them, he notes. Perhaps their observation of the ball’s gradual rotation creates an illusion that the ball itself is moving when it is really just the seams. There is some evidence to back this up. Take a look at this animation. Even though the ball is moving straight down, the changing pattern of shading creates the illusion that it is traveling at an angle. Nathan offers up a fun and effective way to test this hypothesis: Paint the seams of a baseball white (thus removing them as a visual cue) and have hitters take hacks against some knuckleballs. Nathan hasn’t set up such an experiment yet, but he is beginning a study to measure the knuckleball in a wind tunnel to determine why the ball moves so randomly.

Perhaps Nathan’s analysis will give us a better understanding of how physics, possibly with some psychology thrown in, makes Dickey’s knuckleball such an effective pitch. Until then, we can marvel at the fact that one of the slowest hurlers in the majors just received pitching’s most prestigious honor.

Updated 11/15/12 to reflect that Dickey won the award.

 R.A. Dickey image courtesy of slgckgc / Flickr

  • Eric Walker

    What a farrago. First, the knickleball does not “baffle” physicists, which is why its working can easily be described. But second, that description is backward; did not the author (or editors) note the disconnect between “generate spin (up to 50 rotations a second) that makes the ball break, or deviate from a straight-line trajectory” and “When a typical pitcher throws, the ball spins so quickly that those stitches become irrelevant”?

    As the normal pitched ball rotates, the stitches on the top are moving one way and the stitches on the bottom the other; the interaction of those seams with the atmosphere produces differential air flows, hence, by Bernoulli’s Principle, differential pressure, so the ball moves in the direction of the lesser pressure. The pitcher can angle the ball and, to some extent, control the degree of rotation, alloiwng all the wonderful tricks like sliders and curves and cut fastballs.

    The knuckleballer, by holding the ball down to nearly no rotation, leaves its movement subject to randomness rather than the definite, rotation-dominated movement of normal pitches.

    Not terribly baffling.

  • ranjith

    bowlers in cricket have used the seam to swing the ball for more than a century

  • geack

    “one of the slowest hurlers in the majors …” Just to be clear, though – Dickey throws harder than any previous knuckleballer. He can throw in the high 60s or the mid 80s with the same delivery, making him much less a one-trick pony than his predeccesors.

  • Jay29

    I think an issue here might be that Nathan’s data from Pitch F/X, while revolutionary for baseball, is still not precise enough to prove a knuckleball doesn’t ‘dance.’ Pitch F/X might show a knuckler moving in a straight trajectory, but it is still possible that the ball moves enough in flight to be picked up by the batter’s eye and less than the 1-inch error range of Pitch F/X.

  • monkeybotdad

    “I always thought the knuckleball was the easiest pitch to catch.
    Wait’ll it stops rolling, then go to the backstop and pick it up.”
    Broadcaster and former catcher Bob Uecker

  • Jim Abbott

    Hey, where you been buddy?? We miss you!

  • Uncle Al

    Optical dispersion. Red and blue have the most extreme depth of field dispersions for given accommodation. Red will appear too far way versus actual distance. If this is the source of a kuckleball’s effectiveness, then automotive brake lights need re-tinkering, depth of field illusion vs. blinding at night.

  • John

    What Eric Walker says. There is a disconnect between the writer’s assertions that spin on the ball causes movement, in one sentence, and that spin on the ball makes the stitching irrelevant in another sentence.

    Please clarify.

  • Jay Aich

    @ranjith: so have pitchers. Moreover, bowlers bounce the ball. Baseball pitchers do it all in the air. Not only that, in cricket batsmen know the strike zone perfectly (a static wicket), wear padding, and swing a bat equal to about half the size of the wicket. In baseball, the zone is variable; the batters are not nearly as protected, and the bat is miniscule. I don’t know how the sports can be compared.

  • Jay Aich

    Eric Walker’s response is perfect. Moreover, I believe good batters, like Ted Williams, can see the spin on the ball. I know I can at my level (far from the MLB though).

  • geack

    @2. ranjith,

    So have baseball pitchers.

  • Alan Nathan

    Re Eric Walker’s comment (and others that follow): There is no disconnect between the statement that spin causes movement and that the seams are largely irrelevant to the movement on a spinning ball. For a rapidly spinning ball (typical of “normal” pitches), the movement is due to the so-called Magnus force. And the Magnus force exists even on a smooth ball without seams. The seams play very little role on the amount or the direction of the movement. Instead the amount of movement depends on the rotation rate and the direction of movement depends on the axis of rotation (and not on the seam orientation, generally speaking). There are some unusual exceptions, such as the movement that sometimes occurs on Freddie Garcia’s splitter. These are experimental facts, not just theortical predictions. On the other hand, the movement on a knuckleball is not due to the Magnus force but instead is due to changes in the character of the air flow as it passes over the seams. You can read a lot about my own research on the knuckleball at this link:

  • Alan Nathan

    Re Jay29: For the PITCHf/x data I analyzed, the precision of the tracking was better than 0.5″. And the precise statement about the smoothness of the knuckleball trajectory is as follows: Within the precision of the tracking system, the knuckleball trajectory is as smooth as that of normal pitches. Any zig-zag behavior of the knuckleball must be less than a few tenths of an inch or it would have been apparent in the data.

  • Alan Nathan

    Just want to expand a bit on my comment #12. The mechanism I described for the movement of a knuckleball generally does not apply for a normal, rapidly spinning pitch, since the seams are not in any given orientation as the ball rotates. Seams are crucial for how a knuckleball works. Not so for normal pitches.

  • Alan Nathan

    Another comment, re ranjith #2: That is absolutely correct. The so-called “swing” of a cricket ball is due to the interaction of the air with the seam, just as in the knuckleball. The cricket ball has an equatorial seam (unlike the figure-8 pattern of a baseball) and the ball is usually spinning about that seam much like the earth spins on its own axis. Therefore, the orientation of the seam does not change as the spinning ball moves along its trajectory. To get the same effect with the knuckleball in baseball requires that the ball not spin (or spin very slowly).

  • darrell

    Trajectory of the knuckleball is all about Yen and Yang. Picture the stiches of the baseball as the borders of the Yen and Yang symbols inside the circle (baseball). When the stiches are perpendicullar to the direction of wind resistence the ball rotates in the opposite direction until the wind affects the stiches that become exposed in the direction perpendicullar to the turn of the baseballs trajectory. Therefore the up, down, left, and right movements. Watch for the large part of the Yen or Yang of the baseballs seam and you will know which way the ball will move.


Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!


See More

Collapse bottom bar