A measurement of piano playing
How smoothly the melody's loudness changes from note to note.
The same passage — Bach Prelude in C, bars 1–3 — from the recordings that sit lowest and highest on this measurement. Every clip is a note list replayed on one sampled grand at one gain, so the only difference you are hearing is the playing.
Share of melody-velocity variance surviving a five-note moving average.
Every piece in the library, with the range across its recordings. The units mean the same thing in every row, except for basic tempo, which is in each piece’s own beat.
| Piece | Recordings | Lowest | Median | Highest |
|---|---|---|---|---|
| Chopin Étude 10/1 | 31 | 0.25 | 0.46 | 0.70 |
| Ravel Pavane | 33 | 0.49 | 0.56 | 0.67 |
| Träumerei | 31 | 0.41 | 0.57 | 0.71 |
| Schubert Op. 90/3 | 43 | 0.42 | 0.61 | 0.76 |
| Chopin Étude 10/12 | 19 | 0.44 | 0.62 | 0.73 |
| Gymnopédie 1 | 23 | 0.45 | 0.63 | 0.75 |
| Schubert Op. 142/3 | 9 | 0.63 | 0.66 | 0.72 |
| Bach Prelude in C | 27 | 0.47 | 0.67 | 0.86 |
| Ravel Ondine | 61 | 0.55 | 0.69 | 0.82 |
| Mozart K. 545 | 37 | 0.52 | 0.69 | 0.77 |
| Brahms Op. 118/2 | 28 | 0.56 | 0.70 | 0.78 |
| Pathétique II | 47 | 0.64 | 0.71 | 0.78 |
| Nocturne Op. 9/2 | 32 | 0.56 | 0.74 | 0.84 |
| Chopin Ballade 4 | 12 | 0.71 | 0.74 | 0.79 |
| Nocturne Op. 48/1 | 27 | 0.60 | 0.75 | 0.81 |
| Clair de lune | 27 | 0.67 | 0.75 | 0.85 |
| Fantaisie-Impromptu | 20 | 0.69 | 0.80 | 0.88 |
| Rachmaninoff Prelude | 29 | 0.74 | 0.86 | 0.93 |
A Disklavier capture — the pianist’s own key sensors, no microphone — was rendered to audio, transcribed as if it were a recording, and re-measured. The error the transcription step introduces, as a multiple of how far pianists actually differ on this dial, is 2.47. noisy The transcription moves this dial about as far as pianists differ on it, so no claim should rest on where a single recording sits.