A measurement of piano playing
When a passage comes back, how closely its timing repeats the first statement.
The same passage — Pathétique II, bars 29–31 — 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.
Correlation between the beat-level log-tempo profiles of the two passages.
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 |
|---|---|---|---|---|
| Rachmaninoff Prelude | 29 | -0.57 | -0.15 | 0.45 |
| Nocturne Op. 48/1 | 27 | -0.07 | 0.26 | 0.54 |
| Nocturne Op. 9/2 | 32 | -0.26 | 0.33 | 0.84 |
| Brahms Op. 118/2 | 28 | -0.31 | 0.38 | 0.68 |
| Schubert Op. 90/3 | 43 | 0.07 | 0.59 | 0.96 |
| Clair de lune | 27 | 0.52 | 0.72 | 0.91 |
| Mozart K. 545 | 37 | 0.07 | 0.76 | 1.00 |
| Ravel Pavane | 33 | 0.01 | 0.80 | 0.94 |
| Pathétique II | 47 | 0.46 | 0.81 | 0.96 |
| Gymnopédie 1 | 23 | 0.13 | 0.85 | 0.96 |
| Schubert Op. 142/3 | 9 | -0.69 | 0.86 | 0.97 |
| Fantaisie-Impromptu | 20 | 0.67 | 0.87 | 0.95 |
| Träumerei | 31 | 0.60 | 0.89 | 0.97 |
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 0.27. solid This dial survives transcription intact: where one recording sits on it can be argued from.