When Bob Beamon leapt 29 feet, 2½ inches to set a world record at the 1968 Olympics, Scott Cacciola says, his performance was a total anomaly:
He smashed the old record by nearly two feet. It helped that Beamon was jumping in the thin air of Mexico City with a slight wind at his back, but Beamon also benefited from a flash of biomechanical wizardry not even he could replicate.
Experts who have watched the jump on video say Beamon was somehow able to sweep his trail leg across his body at twice his normal rate as he hit the board, creating an explosive whipping action that launched him into the exosphere. Beamon never again broke 27 feet. His record stood for nearly 23 years until Mike Powell eclipsed the mark by two inches in 1991 — a record that still stands.
Video technology has made a major impact on the long jump:
Ramey, who’s studied the sport since the late 1950s when he was an athlete at Penn State, now uses a point-and-shoot camera that records at 200 frames per second, which is seven times more powerful than most video cameras. Though long jumpers only spend about .15 seconds on the takeoff board, his camera is fast enough to capture 30 frames of that sequence alone.
The trouble, Ramey said, is that long jumpers don’t merely want images to study. They want numbers. In the absence of those numbers, jumpers are often left to go on “feel” during training: Does one technique feel more effective than another?
That’s where BMW engineers and their cameras come in:
The distance of a long jump hinges on the moment when a jumper transfers his horizontal velocity (running speed on the runway) into vertical velocity as he leaps off the board. Most elite long jumpers are able to generate a vertical velocity that is about a third of their horizontal velocity — but, generally, the higher that ratio, the longer the jump.
BMW’s research involves using a special “stereo” camera outfitted with two lenses to film athletes as they jump. The camera, which BMW is developing for the purposes of lane-detection systems in its automobiles, turns video into data that is processed through an algorithm on an open-source robotics system. After a jump, the system spits out three crucial numbers on a trackside monitor: A jumper’s horizontal velocity, his vertical velocity as he left the board and his angle of flight.