Click Speed Test

Four details decide whether a CPS number means anything.

1. A monotonic clock. The window is bracketed by performance.now(), which has sub-millisecond resolution and cannot jump when the system clock adjusts mid-run.
2. The measured duration, not the nominal one. A five-second timer fires late by however long the browser was busy, so the division uses the actual elapsed time from your own run. The difference is often small, but using the measured value avoids pretending the callback fired at an exact instant.
3. The starting click counts. The click that begins the test happens inside the window, so this tool counts it rather than silently discarding it. That convention is printed before the run so repeated results remain interpretable.
4. Every primary click counts individually. There is no debouncing and no double-click pairing: if the switch registers, it counts. Right-clicks open the context menu instead and never enter the count.

A short run emphasizes starting timing and a brief burst. A longer run averages more clicks, pacing changes and ordinary variation into one result. That difference in what the window samples is enough to make the readings non-equivalent even on the same device.

The practical consequence: a CPS score is only comparable within its own window length. Compare your five-second runs across days to see whether the number moved; do not compare a one-second personal best to a ten-second one, and do not compare either against a leaderboard that used a different window without checking what that leaderboard actually measured.

Click-speed tools commonly describe several input methods. They are not interchangeable, because each method changes what the counter is measuring:

• Regular clicking — one finger on one button, pressed repeatedly. It is the simplest method to repeat when comparing your own runs.
• Butterfly clicking — two fingers (usually index and middle) alternating on the same button. It should be compared only with runs using the same method.
• Jitter clicking — tensing the forearm so the hand vibrates and the fingertip repeatedly actuates the switch. Stop if the movement is uncomfortable.
• Drag clicking — dragging a finger across the button so friction can produce repeated registrations from one motion on compatible hardware. The result reflects switch behavior and technique together.

Whatever method you use, keep attempts brief and stop if you feel pain, numbness or persistent discomfort.

Three instruments on this bench measure hand speed and none of them produce the same number. The reaction time test measures the delay before movement starts — one stimulus, one response, reported in milliseconds. The aim trainer measures directed movement — getting the pointer onto a target obeys Fitts’s law and takes hundreds of milliseconds per target. This page measures only cadence: how many times per second the switch can be actuated once movement is irrelevant.

A person can be excellent at one and ordinary at the others. Fast twitch response does not imply high cadence, and neither implies precise pointing — which is exactly why they are separate instruments rather than one composite “gaming score”.