Perfecting the Knuckleball

Why is a knuckleball so difficult to throw? It’s a question that major league pitchers will no doubt be asking themselves as they begin spring training. But it’s also a question that inspired a team of fourth-year U of T engineering students to design a machine that could replicate the pitch’s unique motion.

David Sinton, a professor of mechanical and industrial engineering, conceived of the task as a “capstone” project for his graduating students. “Mystery has always surrounded the knuckleball pitch,” he says. “Generating truly repeatable and controllable motion from the knuckleball, mechanically, would make history.”

The knuckleball requires the pitcher to grasp the ball by the finger nails and raised knuckles. Such a grip allows the pitcher to minimize the ball’s spin, making it highly sensitive to variations in the air on its way to the plate (a spinning ball flies through the air more smoothly). Changing the release point or any other variable generally results in a wild pitch – making it incredibly difficult for pitchers to master. R.A. Dickey, the Toronto Blue Jays’ Cy Young Award-winning pitcher, is currently one of only about a dozen professional players who throw it.

By adapting a used pitching machine with motors, sensors, and PVC tubes, Martin Cote, Alex Gordon, Jessica Tomasi and Queenie Yuan (all BASc 2014) developed a prototype knuckleball machine in the department’s fabrication lab. In an attempt to precisely duplicate each pitch, they painstakingly positioned the ball’s seams and used slow-motion cameras and radar guns to monitor its orientation, spin, speed and flight path.

“We could get the ball to curve down to the left every time – so it was somewhat repeatable, says Cote, “but we couldn’t put it into the exact same spot.”

The students have now graduated and gone on to new projects, so they’re not planning to develop their knuckleball machine into a commercial venture, despite its potential. “Alex did a business plan and there seems to be a market,” says Cote. For now, they’re satisfied with having designed a machine that can replicate what only one pitcher in major league baseball can do. For any group of engineers, that must be tantamount to a home run.