Rhythm Test: How Good Is Your Timing?

What this rhythm test actually measures

The test plays a steady click at 100 BPM. The first four beats are a count-in and are not scored. From the fifth beat on, every tap you make is timestamped and matched to the nearest beat, and the test records how far off it was and in which direction.

At the end you get three numbers. Spread is how tightly your taps clustered, in milliseconds, and it is the one that drives your score. Average offset is whether you sat ahead of or behind the click on average. Beats counted is how many of your taps were close enough to a beat to be scored at all, because a tap that lands more than half a beat away cannot be attributed to any particular beat with confidence.

If your offset comes back consistently negative you are rushing, and if it comes back consistently positive you are dragging. Both are habits with identifiable causes and specific fixes, which is the subject of why you rush when you play, and why it gets worse at slow tempos.

Why consistency beats accuracy

Most online rhythm tests score you on raw closeness to the beat. That is the wrong measurement to take through a browser, and here is why: between the click leaving your speaker, travelling to your ears, and your tap being registered by the page, there is a fixed delay made of audio output latency, Bluetooth buffering if you are on wireless headphones, and input handling. On some setups that adds up to a few milliseconds. On others it is well over a hundred.

That delay shifts every tap by roughly the same amount. It makes a metronomically perfect player look consistently late, and it makes a comparison between two people on different devices close to meaningless.

What your score means

The scale is built from the same two thresholds the app uses. A spread of about 25 ms or tighter earns full marks, a spread of about 110 ms or wider earns none, and everything in between scales smoothly across that range.

One reading is a snapshot, not a diagnosis. Tiredness, an unfamiliar device, and simply not having warmed up all cost you points. Run it three times and take the best, which is closer to what you are actually capable of than any single attempt.

Three things that move the number

Subdivide out loud. Most wandering pulses come from counting only the beat. Count the eighths or sixteenths between beats, out loud or in your head, and the gaps stop being guesswork.

Practise at a slow tempo, not a comfortable one. Slow tempos are harder because the gaps are longer and there is more room to drift. If your spread is wide, it will be widest at 60 BPM, which is exactly why that is the useful place to work.

Measure the result, every time. Playing along with a click while never checking the outcome is the plateau most players get stuck on, because in the moment your ear quietly forgives what it just did. Objective feedback after the fact is what breaks that loop, and it is the whole idea behind the feedback loop practice usually misses.

Take the same test with your instrument

Tapping a key measures your timing through a browser. Playing measures your timing. Metro Gnome's Groove Check does the same analysis on your phone but listens through the microphone instead, so you can clap, play a drum, or pick a string and be scored on what actually came out of your instrument. It separates your playing from the click by frequency rather than volume, corrects for your specific device's audio delay with a one-time calibration, then grades the whole session on the same consistency principle described above. It runs entirely on your device and there is no account.

Questions about rhythm and timing tests