Drawing lines on a golf swing video

Every plane line traces back to one 1957 drawing

Ben Hogan and Herbert Warren Wind put a pane of glass into a golf book in 1957. It ran from the ball up to the shoulders, tilted by the golfer's height and his distance from the ball, with his head through a hole in it. Young-Hoo Kwon and Jeff Casebolt's review of the swing plane literature traces the habit of drawing lines on video back to that picture.

What came after is the trouble. Kwon's review concludes that the downswing is not planar and that a single swing plane does not exist, because the clubhead, the hands, the elbows and the shoulders each sweep out their own surface, tilted differently. His alternative, the functional swing plane, is fitted to the clubhead's own path around impact rather than borrowed from a body line.

The shaft line has to choose which plane it means

≈19 degreesBetween the clubhead plane and the trail elbow plane, low handicappers

Stephen Nesbit modelled 84 male and one female amateur golfers and reported that the swing traces two surfaces rather than one: the clubhead sweeps out the first, the hands sweep out the second, and the angle between them runs from 9 to 12 degrees. Draw along the shaft at address and you have committed to one of the two without saying which.

Kwon's laboratory measured those surfaces separately. Twenty-seven right-handed men, fifteen off three or better and twelve off 12 to 18, hit drivers in front of a ten-camera system at 250 Hz. Among the low handicappers the clubhead plane sat at 48.3 degrees, the lead hand plane at 57.9, the trail elbow plane at 67.4. Roughly 19 degrees of spread inside one swing, and your line lands somewhere in it.

The camera is never quite square to the plane

23 cmClubhead's distance from its own swing plane at the top, low handicappers

Where the plane points matters as much as how steeply it tilts, and it does not point along your target line. That study measured direction as well as slope: in the low handicappers the clubhead plane aimed 3.8 degrees to the trail side of the target, while the higher handicap group averaged 0.4 degrees with a spread of plus or minus 6.2. A camera on the target line never sees that plane quite edge-on.

The frame most people pause on makes it worse. At the top of the backswing the clubhead sat 23.2 cm off the plane it would later swing on in the low handicap group, and 36.5 cm off it in the higher one. Whatever the shaft is doing up there, it has not yet reached the surface your line stands for. What being on plane at the top is worth has its own guide.

The spine line is a projection, not your spine

≈9 degreesGap between the flat and three-dimensional side bend at impact, 15 golfers

Aimee Smith and colleagues took one set of marker data from fifteen golfers off handicaps between 1 and 29, driving in front of a thirteen-camera system at 250 Hz, and reduced it twice: flattened onto a plane, which is what a camera does to the world, and computed in three dimensions. The flat version disagreed with the other by 6.54 degrees of trunk flexion at the top and 8.65 degrees of side bend at impact.

Those are the two frames people stop on. The 95 percent limits of agreement at impact ran from about 4 to 22 degrees, which is the honest width of a spine line drawn on a still. A hip line carries the same fault sideways, since a pelvis turned away from the lens reads as a shorter line rather than a rotated one. The box round the head measures travel across the picture only.

Two coaches, one frame, two defensible lines

Line drawing has published protocols, and the protocols contain judgements. The Titleist Performance Institute's test for a flat shoulder plane asks for a line from the base of the tailbone to the back of the neck, then a second line just above the trail shoulder and parallel to the first, and a check on whether the lead shoulder sits above it at the top. Just above is where two coaches part company.

There is a deeper reason to hold any reading loosely. In Kwon's data the steepness of the clubhead plane barely separated the groups: 48.3 degrees in players off three or better, 49.4 in the 12-to-18 handicappers. What told them apart was how the arm and shoulder planes sat relative to that one. No published study has put two coaches in front of the same frame and measured how far apart their lines land.

What a drawn line is still good for

A line makes a poor absolute and a decent relative. Film from the same mark on the ground, at the same height, with the same lens, pause on the same event, and every error described above stays roughly constant from clip to clip. What moves between two of your own videos is then mostly you.

That answers the question golfers actually have, which is whether a change took. It cannot answer the comparison they actually make, against a tour player filmed on another range at an unknown distance. Pebble Swing reads joint angles out of the frames and the backswing-to-downswing tempo ratio out of the motion, and those angles carry the same flattening your drawn lines do. A line holds one thing steady while you change another.