The practical answer: there is no universal millisecond boundary between “tight” and “bad.” In a drum groove, deviations around 20–30 ms are an important transition region: small enough to feel close, but large enough to affect the result or become noticeable in the right context.
StickToClick therefore treats ±25 ms as tight, ±40 ms as normal practice timing, and ±60 ms as loose. Those are useful training windows informed by timing research—not laws of hearing.
Zero milliseconds is not the only good result
When people synchronise movements to a metronome, they do not produce identical offsets on every beat. Research on sensorimotor synchronisation repeatedly finds both small systematic biases and beat-to-beat variation. Performers may also anticipate a click slightly rather than land exactly after its acoustic onset.
That means a single hit at −8 ms is not evidence that you rushed, and a single hit at +12 ms is not evidence that you dragged. The useful question is whether a pattern develops across the take.
What the timing number measures
StickToClick compares the timestamp of the MIDI hit with the expected onset in the score. A negative value is early; a positive value is late. Before scoring, the app applies the latency correction measured for your kit, so the number is intended to describe your playing rather than the fixed delay in the hardware path.
Calibration matters because electronic kits, USB interfaces and audio devices do not all report events at the same point in their signal path. Bluetooth and AirPlay add variable output latency and clock behaviour, so a wired audio path is the reliable choice for millisecond practice.
How to interpret the three windows
Tight ±25 ms
A demanding click-lock target. Small deviations can still shape feel, but hits inside this window are close enough that many will not stand out as discrete timing errors in a full groove.
Normal ±40 ms
A practical everyday training window. Hits near its edge may sound or feel loose—especially on exposed snare and hi-hat parts—but the window is useful while building control.
Loose ±60 ms
A learning and note-placement window. It helps distinguish the intended note from a clear miss without pretending that every accepted hit is perceptually locked to the click.
The wider match window used internally is different. It lets the app decide which notated note a hit belongs to. A hit can be matched correctly and still be labelled early or late.
Perceptible is not the same as unacceptable
Perception changes with the instrument, sound envelope, tempo, subdivision, arrangement and listener. Percussive sounds have sharp attacks, so their relative timing is exposed. Dense backing tracks can mask small differences; isolated click-and-pad practice reveals them.
In a controlled drum-pattern study, perfectly quantised versions received the strongest groove-quality ratings, while imposed microtiming deviations of 15 and 25 ms reduced ratings as the displacement increased. Early shifts were generally judged more harshly than equivalent late shifts, and the affected drum voice also mattered.
That is why StickToClick uses symmetrical windows for clarity, but still shows the signed offset. The verdict tells you whether a hit met the selected training target; the sign tells you which way your timing is leaning.
Read the take, not just the hit
1. Accuracy
How many expected notes fell inside the selected correct window? This is the simplest progress measure, but it does not explain the shape of the errors.
2. Direction
Are most non-correct hits early or late? A stable bias is different from random scatter. Ten hits around −28 ms suggest a repeatable rushing tendency; alternating between −55 and +55 ms suggests inconsistency.
3. Consistency
A drummer whose hits cluster closely around +12 ms is often more controlled than one whose average is 0 ms only because large early and late errors cancel out. Tightness is fundamentally about spread as well as average position.
4. Musical location
Look for repeated trouble on a limb, drum, subdivision or transition. A late kick after a fill is more actionable than an isolated late hit somewhere in a long take.
Four concrete examples
−6 ms: effectively locked for ordinary practice. Do not chase zero at the expense of relaxed playing.
+23 ms: inside Tight, near the edge. Worth watching only if many similar hits form a pattern.
−37 ms: correct on Normal but early on Tight. This may be audible in exposed material and is a useful improvement target.
+82 ms: clearly late for click-based playing. It may still be associated with the intended note by the wider match window, which is why the app can show a meaningful “late” verdict instead of calling it an unrelated extra hit.
What the research does—and does not—justify
The literature supports measuring both average asynchrony and variability, recognises that musicians often anticipate metronome events slightly, and shows that small drum microtiming changes can alter perceived groove. It does not establish one universal pass mark for every drummer, tempo, style and mix.
The StickToClick windows are therefore an engineering and teaching interpretation of the evidence. Tight is intentionally demanding; Normal is intentionally useful; Loose is intentionally forgiving. Change the level as your purpose changes.
Sources
- Frühauf, J., Kopiez, R. & Platz, F. (2013). “Music on the timing grid: The influence of microtiming on the perceived groove quality of a simple drum pattern performance.” Musicae Scientiae, 17(2), 246–260. doi:10.1177/1029864913486793
- Repp, B. H. (2005). “Sensorimotor synchronization: A review of the tapping literature.” Psychonomic Bulletin & Review, 12, 969–992. doi:10.3758/BF03206433
- Repp, B. H. & Su, Y.-H. (2013). “Sensorimotor synchronization: A review of recent research (2006–2012).” Psychonomic Bulletin & Review, 20, 403–452. doi:10.3758/s13423-012-0371-2
Put the milliseconds on the score
StickToClick scores every expected drum note against your chart, with latency calibration and Tight, Normal or Loose timing.
Get StickToClick free — beta for macOS