How the test works

You leave the ground at some speed v, rise until gravity stops you, and fall back for the same time. So the time in the air is t = 2v/g, and the height your hips rise is h = v²/2g. Put together: h = g·t²/8, with g = 9.81 m/s².

A 30 cm (11.8 in) jump keeps you in the air for about 495 ms. Every result also shows your take-off speed, v = g·t/2.

Jump mats and validated phone-video apps use this same flight-time method [1]. In a lab, a force plate measures take-off speed directly from how hard you push, which makes it the reference [4].

Take-off is the first frame where both feet are off the ground. Landing is the first frame where a foot touches down again. The flight time is the time between those two frames.

Both marks are late by up to one frame, in the same direction: the real take-off happened somewhere between the marked frame and the one before it, and so did the real landing. On average the two errors cancel, which is why the result's precision is plus or minus one frame.

One frame of error changes the height by g·t/4 times the length of a frame. Higher jumps and slower video give bigger bands. Slow motion from your camera app is the most precise way to test.

A study that filmed jumps at 1,000 fps and cut the video down to lower rates found a timing error of 1.4% of the height at 120 fps and 0.7% at 240 fps, with little to gain above that; it recommends 240 fps [3]. Those are measured spreads, smaller than the one-frame limits in the table.

A video's timestamps say when each frame was filmed, unless the phone has stretched them to play slow motion slowly. The tester reads every frame's timestamp. When a video mixes real-speed frames with stretched ones, it counts frames at the camera's rate instead. When a whole video is stretched, the flight looks impossibly long, and the tester asks what rate you filmed at.

On iPhone, slow-motion videos picked from Photos arrive with the slow motion baked in, and the tester handles that. A video picked from Files arrives exactly as the camera made it.

• —Tucking your knees or landing with bent legs. You stay in the air longer than you rose, so the height comes out too high: typically by 0.5–2 cm, and more with arm swing [2]. Land with straight legs, toes first, on the spot you took off from.
• —Feet out of the picture, or a camera so high you can't see the gap under the shoes. Put the phone on the floor or at knee height, 2–3 m away.
• —Marking the wrong frames. Zoom in on the feet and step one frame at a time.
• —A shaky camera. Lean the phone on something.

We are comparing jumps recorded live in the tester at 60 fps with the same jumps filmed in 240 fps slow motion. The results will be published on this page. Until then we make no claim beyond the arithmetic above.