CPS Clicking Techniques, and What the Counter Actually Measures

The counter does not measure skill. It measures events the mouse reported. That distinction explains every surprising number on a clicks-per-second test, including the ones that look impossible.

Clicking-speed tests attract a lot of technique advice, most of it recycled without explanation. Here is the mechanism first, then what follows from it.

What the counter records

Every pointer event on the target area is timestamped while the timer runs. Two figures come out of that list:

average CPS = total events / elapsed seconds
peak CPS     = max events inside any rolling 1-second window

Peak is the one people misread. It is not a separate measurement of speed — it is the maximum density over a single second anywhere in the run, which means a strong opening burst alone can produce an impressive peak regardless of how the rest went.

A worked example

Consider a ten-second run of 54 clicks that opens hard and then settles. The timestamps below are illustrative rather than recorded, but the arithmetic is exactly what the counter runs.

A peak 54 with an average of 5.4 is a recognisable pattern: the opening second carried the run. Nothing about the peak is dishonest, but reading it as "this is my clicking speed" would be a mistake — the average is the figure that describes the run.

You can reproduce it: any burst faster than the sustained pace raises peak without moving the average much. Two runs with identical averages can have very different peaks.

The three named techniques

Regular clicking

One finger, one mouse button, one event per press. Most people land between 4 and 7 CPS here. This is the only technique whose score reflects clicking ability rather than hardware behaviour.

Jitter clicking

The forearm is held tensely and the wrist oscillates, so the mouse itself vibrates. If the switch is debounced generously, that vibration registers as several distinct events from a single intended press.

The honest framing: the high number is real in the sense that the hardware reported those events, but it is a property of the switch's tolerance as much as of the person. It also fatigues the forearm quickly, so sustained jitter numbers collapse well before regular clicking does.

Butterfly clicking

Two fingers alternate on a single button. Both hands rest on the mouse, which is the source of most of the gain — there is simply less travel per event.

Unlike jitter, this produces genuine distinct presses, and it can roughly double a regular score. It is also the technique most likely to produce accidental drags and misfires on a bad mouse surface, which show up in the count as extra events.

Drag clicking

A finger is dragged back and forth across the button so it bounces under friction, registering many events per stroke. It generally produces the highest peaks of the four.

The catch is that it depends heavily on the button's physical design, and it can be recorded as a drag-and-drop rather than a series of clicks by software that inspects event types. If a counter reports very high drag numbers, that is a reason to be curious about what it is accepting.

What actually determines your score

In rough order of influence, and mostly not under your control:

1. Button debounce interval. The single biggest factor. A mechanical switch with a short double-click window registers more events than a membrane switch.
2. Whether the OS is filtering. Some pointer acceleration or click-lock settings change what reaches the page.
3. Hand position and travel. Butterfly clicking wins here by reducing it.
4. Surface friction. Affects drag clicking more than anything else.
5. Stamina. Governs the average far more than the peak, because it decides whether you can hold pace for the full duration.

That ordering is why two people with identical technique can post numbers 40% apart, and why a CPS figure is close to meaningless outside the exact setup that produced it.

Are these allowed?

That is not our call. Several competitive games and anti-cheat systems prohibit multi-click techniques explicitly, and using them there risks bans that a personal benchmark never will. Check the rules of anything competitive before taking a technique from this page.

The average and peak formulas quoted are this site's own, frompublic/test-scripts/cps-test.js. The timestamps in the worked example are constructed for illustration — they are not a recording of a real person's run, and no user data is collected anywhere on this site.