What Is a Good Reaction Time?

It is roughly the middle figure from laboratory studies that used purpose-built response buttons. It has been repeated so often that it floated free of the conditions that produced it.

Those conditions matter. A purpose-built button has no screen lag, no operating system queue and no browser sitting between the signal and the timestamp.

What you are reacting to matters too. Sound beats light by 30 to 40 ms. Touch is faster still. And a signal you can half predict beats one you cannot. A single number quoted without any of that is a number about nothing in particular.

Four things, in rough order of size.

• Your screen. A 60 Hz monitor can take up to 16.7 ms to show a change, plus however long the panel itself needs. Televisions used as monitors often add far more.
• Your mouse or keyboard. How often it reports, and how long it waits out the switch bouncing, cost a few milliseconds each.
• Your operating system, which decides when to pass the press along.
• Your browser, which hands that press to the page at its next opportunity.

Those four together usually total 30 to 80 ms. So your real reaction time is faster than the number you see.

We do not take a guess off your score. The true figure depends on hardware we cannot see, and a flattering guess would make the number less honest rather than more accurate. The methodology page sets out exactly what we measure and where we stop.

Your trials are not spread evenly. There is a hard floor on the fast side that nobody crosses, and a long straggle of slow trials above it.

That is why an average is a poor summary, and why every reaction test on this site reports a median instead. One inattentive trial in five moves an average by 80 ms. It moves a median by almost nothing.

It is also why your own spread is worth knowing. Two people with a 265 ms median can differ enormously in how much they wobble, and the wobblier one is drifting more often. The psychomotor vigilance task is built on exactly that. It ignores your average and counts your lapses.

Notice how the list starts. The two biggest factors have nothing to do with your brain.

• Hardware, by 30 to 80 ms.
• Sleep, by tens of milliseconds when you are meaningfully short.
• Age, by roughly 50 to 60 ms across four decades of adult life. Average reaction time by age has the curve.
• Time of day, by 10 to 20 ms between your worst and best hours.
• Caffeine, by a few milliseconds.
• Practice at this exact task, by 10 to 20 ms over the first few runs and very little after that.

It is not a measure of intelligence. Across large groups, faster reaction times go with slightly higher scores on thinking tests. The link is far too weak to say anything usable about one person.

It is not a measure of athletic ability either. Sport depends far more on reading the play, spotting patterns and moving well than on raw detection speed. Elite athletes are not dramatically faster than anyone else on a click test. They are dramatically better at knowing what is about to happen.

And it is not a health check. A slow score means your equipment is slow, or you are tired, or the task is new to you, long before it means anything else. If your reaction time has changed in a way that worries you, see a doctor rather than a stopwatch.

Outside the internet, response speed is used as an instrument rather than a score. Three fields rely on it.

Sleep research uses it because lapses pile up with sleep debt in a steady, measurable way. People's own sense of how tired they are does not.

Drug research uses it because sedating drugs slow simple responses at doses well below the point where anyone looks impaired.

Ageing research uses choice reaction time because it changes more steeply than the simple kind. It is also easier to measure reliably than most things in psychology.

In all three the value comes from measuring the same person on the same equipment, over and over. It does not come from placing someone against a population table. Borrow that habit. The gap between two of your own scores is a real measurement. The gap between your score and a stranger's is mostly a gap between two computers.

A number you can repeat. Hit the same median across three runs on one machine and that number means something. A lucky 190 ms you saw once does not. Only the steady figure will tell you anything when you take the test again in six months.

The reliable gains are all boring. Sit properly. Rest your finger on the button instead of hovering. Use a mouse rather than a trackpad, and close the tabs doing background work.

Then stop after three runs. Past that, tiredness costs you more than concentration wins. Try the choice reaction time test instead, where the interesting number is not your speed but the price you pay for deciding.