Reflex

The test

Reflex measures the gap between a visual signal appearing and your finger responding to it. Tap once to arm a round. The panel turns red and holds there for an unpredictable interval — anywhere between one and five seconds — before switching to green. The instant it turns green, tap. The number you see is your reaction time in milliseconds.

The wait is deliberately randomised. If the delay were fixed you would simply learn its rhythm and tap on a count, which measures timing rather than reaction. Tapping before the green appears is a false start: the round does not score, and you repeat it. Five valid rounds make a complete test.

Reading your score

Your headline result is the mean of all five rounds, which smooths out the single lucky tap and the single distracted one. The individual round times are listed underneath so you can see your consistency — a player averaging 260ms with every round between 250 and 270 has genuinely steadier reflexes than one averaging 260ms across a spread of 190 to 340.

For calibration: most adults land between 200 and 300 milliseconds on a visual test like this one. Anything under 200ms is genuinely fast and typically shows up in people who play action games regularly. Under 150ms is exceptional and worth double-checking for accidental early taps. Above 350ms usually points to fatigue, a slow display, or a touchscreen adding its own input latency rather than anything about you.

How the percentile works

After the fifth round Reflex places you on a percentile curve — "faster than 82% of players", for example. That ranking is computed locally against a fixed statistical distribution modelled on published human visual reaction-time research, not against a live leaderboard of other users. No scores are uploaded and no personal data leaves your device, so the ranking is stable, private and available instantly even offline.

Because the distribution is fixed rather than crowd-sourced, your percentile means the same thing today as it will in a year. It is a benchmark, not a competition.

Getting a faster time

Hardware genuinely matters here. A 120Hz phone screen can show the green flash several milliseconds sooner than a 60Hz one, and a wired mouse typically beats a Bluetooth trackpad. Close background tabs before testing — a busy CPU adds real latency to the browser paint.

Technique matters more. Rest your finger lightly in contact with the screen or button rather than hovering above it, so the tap requires no travel. Watch the centre of the panel with a soft, unfocused gaze rather than staring hard at one point; peripheral vision detects sudden brightness changes faster than foveal vision does. And do not try to anticipate the switch — predictive tapping produces false starts far more often than it produces record scores, and the randomised delay is specifically designed to punish it.