Cricket bowling speed calculator for phone video

A free online bowling speed meter, no app to install

This calculator is a free bowling speed meter that runs in your browser. Enter the frame count and frame rate from your own video and the distance the ball travelled, and it returns km/h and mph with the ±1-frame error band. There is nothing to install.

How to count frames on your phone

Film from side-on, square to the pitch, with the phone on a tripod or wedged against something solid, and wide enough to see both the bowler's hand at release and the batter's stumps. From behind the bowler the ball travels almost straight away from the lens, which makes the arrival frame hard to judge.

Use slow motion. On iPhone, Slo-mo records at 240 fps from the back camera by default (Settings > Camera > Record Slo-mo); Google lists slow motion up to 240 fps for recent Pixel phones, and most Android phones have a similar mode. Enter the rate the clip was captured at — 120 or 240 — not the rate it plays back at.

Open the clip in an editor or player that steps one frame at a time. The release frame is the first frame in which the ball is clearly out of the fingers. The arrival frame is the first frame in which the ball reaches the line of the batter's stumps — or the bat, or the bounce, if that is what you are measuring to. Frames = arrival frame number minus release frame number.

If your app shows timestamps rather than frame numbers, subtract the two times and use speed = distance ÷ time directly. Check that the timestamps are real time: slowed playback time on a 240 fps clip would make the ball look eight times slower.

If you have a time in seconds instead of frames

Speed = distance ÷ time works with any time measurement. A ball that covers 17.5 m in 0.45 seconds averages 38.9 m/s: multiply by 3.6 for 140.0 km/h, then multiply the km/h figure by 0.621371 for 87.0 mph. Use the distance the ball actually travelled in that time, and check that the times are real time, not slowed playback.

What distance to use

The pitch is 22 yards (20.12 m) from stumps to stumps under Law 6, but no delivery travels that far before reaching the batter's stumps. The popping crease is 4 ft (1.22 m) in front of the bowler's stumps under Law 7, and for a fair ball some part of the front foot must land behind it under Law 21. The ball then leaves the hand ahead of that front foot.

How far ahead depends on the bowler. A Loughborough study of junior seam bowlers measured release at about 30% of the bowler's height in front of the front heel, with the heel landing about 0.35 m behind the crease. For a 1.8 m adult whose heel lands just behind the line, that puts release about 1.76 m in front of the bowler's stumps and 18.4 m from the batter's; land 0.35–0.5 m further back and it is 18.7–18.9 m. The calculator defaults to 18.5 m, and you can override it.

Distance matters less than frame rate: being 0.3 m out changes the answer by 1.6%. Using 20.12 m, as some online calculators and guides do, inflates every reading by about 9%.

On a junior or shortened pitch, start from that pitch's length and subtract about 1.3–1.8 m. Junior pitch lengths are set by each country's governing body (Law 6.6), so use the length you actually bowled on.

Why your result reads lower than a speed gun

A radar gun reports the ball's speed at or just after release, when it is fastest; broadcast figures such as the 161.3 km/h record are release speeds too. Researchers record that speed at the instant of release; one study defines it as the peak speed between release and the bounce.

A frame count gives the average speed over the whole path. The ball slows all the way down the pitch through air resistance and loses more speed when it bounces, so the average is always below the release speed. How far below depends on the length, the pitch and the ball: two published estimates, which disagree about the bounce, are compared in how a cricket ball slows from hand to bat. We have not found a published measurement we trust enough to subtract, so the calculator applies no correction.

In practice: treat your figure as a conservative estimate, and compare it only with other frame counts measured the same way. Counting to the bounce rather than the stumps removes the loss at the bounce, if you know where the ball pitched.

Accuracy and limits

Frame rate sets the precision. Neither release nor arrival happens exactly on a frame, so the true time can be up to about one frame either side of your count. The calculator shows that band: the speed at frames + 1 and at frames − 1.

The table shows what that means for a 120 km/h delivery over 18.5 m. At 30 fps the ball moves about 1.1 m between frames, and one frame of doubt is worth roughly ±6% — anywhere from 111 to 125 km/h. At 240 fps it moves 0.14 m between frames and the band shrinks to under ±1%.

Other limits. If your phone or app creates extra slow-motion frames by interpolation, those frames are generated rather than captured and add no timing information. A rate shown as 120 fps may really be 119.88, which changes the answer by 0.1%. Motion blur makes the release frame a judgement call, so pick it the same way every time. And the radar you compare against has its own tolerance: one research team quotes ±3% for a common sports radar.

Use the result to track your own progress under the same set-up, not as a certified personal best.

When to use Stance instead

Counting frames gives one number per delivery and depends on a distance you have to assume. Stance tracks the ball automatically and calibrates distance from the stumps in the frame — 71.12 cm tall and 22.86 cm wide under Law 8 — so there is no distance to guess and no frames to count. For the types of phone app that track the ball and what each one measures, see how to choose a cricket analysis app. If you would rather mark frames in an app, our guide to bowling speed meter apps explains how each type measures.

Because the ball is tracked through the delivery, the same clip also gives line and length. The same physics still applies: frame rate sets the precision and camera angle sets what can be measured, so film side-on at 60 fps or higher with the stumps in view. Stance is in closed beta; you can join the waitlist.