What is a good CPS score?

Normal clicking

Ordinary clicking — one finger, moving from the knuckle — produces about 5 to 7 clicks per second for most people. The median ten-second score across large public datasets sits near 6.9, which is a useful anchor because ten seconds is the format almost everyone reports.

Above roughly 8 clicks per second with a single finger you are already unusual, and beyond about 10 you are at the mechanical limit of one finger pressing a switch. Everything above that involves doing something different with your hand.

The technique ranges, and where they come from

Jitter clicking tenses the forearm into a controlled tremor and lets the vibration press the button; it is commonly reported at 10 to 14 clicks per second. Butterfly clicking alternates two fingers on the same button, so each finger works at half the total rate, and reaches 15 to 25. Drag clicking runs a finger across the button so friction produces a burst of contacts, and registers far higher in short bursts.

Be clear about what those numbers are. No formal study has measured clicking techniques. Those ranges are what the Minecraft community consistently reports and moderates against, which makes them a reasonable consensus and not a measurement. We would rather tell you that than dress them up.

The one hard number in this area is a record rather than a range: 830 mouse clicks in 30 seconds, set in 2010 under rules that required video evidence and forbade auto-clickers. That is 27.7 clicks per second sustained for half a minute.

The Minecraft mechanic nobody mentions

Java Edition 1.9 introduced an attack cooldown, and it changed what clicking speed is worth. A sword has an attack speed of 1.6 attacks per second, meaning a full-power swing every 0.625 seconds. Damage scales with how far the cooldown has recharged, following a curve where a hit at zero charge deals only 20% of the weapon’s damage and a hit at half charge deals a little over 40%.

The consequence is blunt: in 1.9-and-later combat, clicking faster than about 1.6 times a second with a sword actively reduces your damage output. A twenty-clicks-per-second butterfly clicker is doing less damage than someone timing their swings.

Clicking speed still decides fights in 1.8-style combat, which is what most PvP servers run, and that is why the whole CPS category exists. But if you are playing modern Java survival or any server on current combat mechanics, the click test is a game rather than training.

What the rules actually ban

The commonly repeated claim that servers cap you at 15 or 20 clicks per second does not appear in any rule text we could find. What major servers ban is automation: auto-clickers and macros, at any speed, classified alongside cheating rather than alongside fast play. Rebinding keys, including to extra mouse buttons, is explicitly allowed.

That distinction explains something that confuses people. Anti-cheat systems look for the signature of a machine, not for speed, which is why an auto-clicker set to a leisurely two clicks per second gets caught while a human clicking at fifteen does not. Human hands wobble by fifteen percent or more even at full speed; a script wobbles by well under two. We use exactly that test to decide whether a run enters our shared statistics.

Duration changes everything

A one-second burst reads two to four clicks per second higher than the same person’s ten-second average, because fatigue has not appeared. A sixty-second test reads one to two lower. Comparing scores across durations is meaningless, which is why each duration here is its own page with its own percentile table.

The per-second chart after each test is the part worth looking at. A flat chart means your limit is technique; a falling one means it is stamina, and the fix for those two problems is not the same.