How we time swings to 1/100 of a second
Elite Tempo publishes tempo figures that disagree with the number almost everyone quotes. Rory McIlroy's 2014 PGA Championship driver comes out at 2.45:1, not the “textbook 3:1” you will find everywhere else. A claim like that is only worth making if you also show your working, so here it is — including where it can be wrong.
The method
Find footage of the actual shot
Not a range swing, not a montage — a specific shot from a specific round, where the camera holds a clean view from takeaway through impact. Every measurement in the library links to the footage it was read off, so the source is checkable rather than asserted.
Mark takeaway, top and impact
Three marks, placed by hand, stepping through the video frame by frame. Takeaway is the first frame the clubhead moves away from the ball. Top is the frame the club stops going back. Impact is the frame the clubface meets the ball.
Derive everything from those three marks
Backswing is top minus takeaway. Downswing is impact minus top. The tempo ratio is backswing divided by downswing, and the total duration is impact minus takeaway. There is no smoothing, no model and no fourth number — which is why the marks themselves are published on every page.
Worked example
Tiger Woods, 2000 The Open Championship, driver. Takeaway at 0s, top at 0.8238s, impact at 1.047s. Backswing 0.82s, downswing 0.22s, so the ratio is 0.82s ÷ 0.22s = 3.69:1 over 1.05s.
Note that the marks carry more decimal places than the published figures. That is deliberate: rounding the timestamps before dividing them drifts the answer, and not by a little. Round this swing's marks to three decimals and you get 3.70:1; round to two and you get 3.57:1. The full-precision marks give 3.69:1.
Where the 3:1 rule comes from
The familiar figure comes from counting frames at 30 frames per second: 24 frames back, 8 frames down. That is a genuinely useful teaching tool — it is easy to count and it puts most golfers in roughly the right place. But counting whole frames rounds every measurement to the nearest thirtieth of a second, and on a downswing lasting about a quarter of a second, one frame either way moves the ratio by a lot.
Timed against the footage directly, the 16 shots in the library spread from 2.02:1 to 4.26:1. 3:1 is a reasonable average of that spread and a poor description of any single swing in it.
What this method cannot do
Marks are stored to 1/100 of a second, but the honest uncertainty is larger than that, and it comes from the video rather than the timing. Broadcast footage varies in frame rate, a clubhead moving at speed smears across a frame, and the exact moment the club stops going back is not always a clean edge. Because the downswing is the shorter of the two intervals, whatever uncertainty exists lands hardest on the ratio. Higher-frame-rate footage tightens this; nothing eliminates it.
Two other honest limits. Each figure describes one swing on one day, not a player's career average — tempo moves with club, lie, nerves and intent. And a camera angle that is not square to the swing can make the top of the backswing genuinely ambiguous by a frame.
This is why the raw marks sit on every page rather than only the tidy ratio. If you disagree with where a mark was placed, you can see exactly which one and by how much it would move the answer.
Common questions
How is golf swing tempo measured?
Three marks are read off video: takeaway, the top of the backswing, and impact. Backswing time is top minus takeaway, downswing time is impact minus top, and the tempo ratio is the first divided by the second. Total duration is impact minus takeaway. Nothing else goes into it, which is why publishing the three marks is enough for anyone to check the arithmetic.
Why do Elite Tempo's numbers differ from the 3:1 rule?
Because 3:1 is an average, not a measurement. It comes from counting frames at 30fps — 24 frames back, 8 down — which rounds every swing to the nearest thirtieth of a second and lands most of them on or near 3:1. Timing the same swings against the footage directly gives a spread rather than a single number.
How accurate are the measurements?
Precision is bounded by how clearly the footage shows the moment. Marks are stored to 1/100 of a second, but the real uncertainty is larger than that and comes from the video: how many frames per second it was shot at, how much motion blur there is on a clubhead moving at speed, and whether the camera angle makes the top of the backswing unambiguous. The downswing is the shorter of the two intervals, so any uncertainty affects the ratio there most. The marks are published alongside every measurement so a reader can judge them rather than take them on trust.
Is the timing done by AI?
Not for the tour library. Every shot in it was marked by hand against the footage, because broadcast video changes camera angle, frame rate and shutter speed shot to shot, and an automatic detector trained on clean side-on video does poorly on it. Swings you record in the app are a different case: those are filmed on one camera in good conditions, and the app finds takeaway, top and impact in them automatically.
See all 16 measurements
Every swing in the library, with its marks, its ratio, and the footage it was timed from.