Swing eighths are not 2:1 — what the research actually shows

Open any jazz method book and you will find swing eighths described as a 2:1 ratio: the first eighth gets two-thirds of the beat, the second gets one-third. This is the textbook definition. It is also, when measured against actual professional jazz performances, almost completely wrong.

The empirical research on swing-eighth performance — much of it from Anders Friberg and Andreas Sundström’s 1999–2002 studies of professional jazz drummers, with extensions through the 2010s and 2020s — shows that the swing ratio varies substantially with tempo, with player, with section of the ensemble, and with role. Slow tempos can produce ratios of 3.5:1 or even higher; fast tempos collapse the ratio to nearly 1:1 (straight eighths). The 2:1 ratio is not the empirical center of the distribution; it is a convenient pedagogical fiction.

Understanding what swing actually is — and isn’t — has practical consequences for how to learn to play it, hear it, and train an ear-training app to recognize it.

The Friberg & Sundström finding

In 1999, Friberg and Sundström at KTH Stockholm presented analysis of recordings of four professional jazz drummers playing brushes on a snare at varying tempos. Their finding was unambiguous: drummers’ swing ratio varied gradually and substantially with tempo, from roughly 3.5:1 at slow tempos (around quarter = 100 BPM) to approximately 1:1 at very fast tempos (around quarter = 300 BPM) ^[1].

Their follow-up work (Sundström & Friberg 2002) extended the finding across more drummers and more tempos and added a critical second observation: the absolute duration of the short note in the long-short pattern stayed roughly constant at about 100 milliseconds across all medium-to-fast tempos. As tempo increased, the long note shortened to maintain the constant 100ms short, until the long and short notes converged in duration and swing collapsed to straight eighths ^[2].

This is a striking result. It suggests that swing is governed not by an abstract ratio but by a perceptual or motor constraint on how short the second eighth can be. Below ~100ms, the second eighth becomes too brief to register as a distinct articulation, and the player adjusts the long note rather than letting the short note get any shorter.

What other studies have added

Subsequent empirical work has refined and complicated the picture without overturning it.

Microtiming deviations beyond ratio. A 2022 paper in Communications Physics (Datseris and colleagues) found that systematic downbeat delays — the soloist or rhythm section playing slightly behind the beat by a few tens of milliseconds — are a key component of the perceived swing feel. Slightly delayed downbeats and synchronized offbeats enhance the swing sensation independently of the long-short ratio ^[3].

Soloist vs rhythm section. The same line of research found that soloists tend to play behind the beat of the rhythm section and at a lower swing ratio (i.e., closer to straight) than the drummer, in order to synchronize their offbeats with the ride cymbal’s offbeats. So within a single ensemble performance, different musicians may simultaneously be using different swing ratios — and this is part of what makes the ensemble feel cohesive ^[4].

Magnitude of variation is jazz-specific. Microtiming deviations in jazz are larger and more systematic than in rock or other Western popular genres ^[5]. This is consistent with the practitioner intuition that “feel” or “groove” is partly an artifact of microtiming — the deviations from strict mechanical alignment are not noise, they are signal.

Common rhythmic patterns across performers. Friberg and Sundström and later Honing and colleagues have argued that, despite the wide variation in swing ratio, certain patterns of variation (the curve of swing ratio against tempo, the constant-short-note-duration finding) are surprisingly consistent across professional jazz drummers. There is a style-level structure to swing, even if there is not a single ratio ^[6].

Why this matters for learners

A learner who reads the textbook 2:1 ratio and tries to perform it mechanically — by playing eighths as triplet quarter-eighth — will sound stiff and unconvincing at most tempos. The reasons:

At slow tempos, the actual swing ratio is closer to 3:1 or higher. Triplet-feel eighths (2:1) sound rushed.
At medium tempos (around quarter = 160), 2:1 is roughly accurate, but the absence of microtiming (the slight downbeat delay) still makes the result sound mechanical.
At fast tempos, the actual swing ratio collapses toward 1:1. Trying to maintain a 2:1 feel at quarter = 240 produces eighths that are too uneven and that drag.

The pedagogical move that works, supported by the empirical literature: listen to recordings at the tempo you are practicing, and match what you hear, not what the chart implies. The professional swing literature documented by Friberg, Sundström, and the subsequent generation of microtiming researchers gives concrete numbers, but the right practice protocol is still imitation — match the recordings, not a ratio.

Implications for ear-training apps

A few practical consequences if rhythm-perception lessons are ever built into an ear-training app.

Synthesizing “swing” with a fixed 2:1 ratio across all tempos is wrong. A swing-eighth quiz that uses 2:1 swing at quarter = 100 will sound less swung than a real recording at the same tempo. A swing-eighth quiz at quarter = 280 with 2:1 swing will sound too swung, almost shuffle-like. The right model is the Friberg-Sundström curve: swing ratio decreases roughly linearly with tempo.

Microtiming deviations should be included in synthesized examples, not avoided. A perfectly quantized swing eighth pattern is not what swing actually sounds like. Even small downbeat delays (10–30ms) make synthesized swing examples sound much more authentic.

Ear-training tasks around swing should focus on hearing the amount of swing, not just identifying its presence. A useful drill: given an excerpt, place the swing on a continuum from “straight” to “shuffle.” This is a more practically useful skill than the binary “is this swung or straight?”

Use real recordings whenever possible. As with odd-meter perception, the deepest fluency comes from extensive listening to authentic performances rather than from synthetic examples. An app that pairs short recorded excerpts with timing-discrimination tasks is doing something the textbook approach cannot.

A note on what swing is for

Beyond the technical question of ratios and microtiming, there is a phenomenological question of why swing exists at all. The current best answer, drawn from the embodied-cognition and beat-induction literatures: swing increases the perceived groove of music, which is the felt urge to move with it. The microtiming deviations make the pattern slightly less metronomic, which paradoxically makes the beat more salient — the listener’s brain has to actively predict and entrain rather than passively receive.

This connects directly to the broader rhythm research discussed in Konnakol and the beat-induction literature: rhythm perception is an active prediction process, and the rhythms that engage it most are the ones that give the prediction system something interesting to do. Perfectly mechanical eighths give the prediction system nothing to do. Swung eighths, with their tempo-dependent ratios and their characteristic microtiming, give it exactly the kind of work it is built for.

References

Friberg, A., & Sundström, A. (1999). Jazz Drummers’ Swing Ratio in Relation to Tempo. Acoustical Society of America / EAA / DAGA '99 conference. Press release: https://acoustics.org/pressroom/httpdocs/137th/friberg.html. The original experimental finding of swing ratios from 3.5:1 at slow tempos to ~1:1 at fast tempos. ↩︎
Friberg, A., & Sundström, A. (2002). Swing ratios and ensemble timing in jazz performance: Evidence for a common rhythmic pattern. Music Perception, 19(3). https://online.ucpress.edu/mp/article-abstract/19/3/333/61900/Swing-Ratios-and-Ensemble-Timing-in-Jazz. ResearchGate: https://www.researchgate.net/publication/238657386_Swing_Ratios_and_Ensemble_Timing_in_Jazz_Performance_Evidence_for_a_Common_Rhythmic_Pattern. The constant-short-note (~100ms) finding. ↩︎
Datseris, G., et al. (2022). Downbeat delays are a key component of swing in jazz. Communications Physics. https://www.nature.com/articles/s42005-022-00995-z. ResearchGate preprint: https://www.researchgate.net/publication/355610766_Downbeat_delays_are_a_key_component_of_the_swing_feel_in_jazz. ↩︎
The soloist-vs-drummer differential swing ratio is documented in the same line of work. See: Friberg, A., & Sundström, A. Preferred swing ratio in jazz as a function of tempo. https://www.semanticscholar.org/paper/Preferred-swing-ratio-in-jazz-as-a-function-of-Friberg-Sundström/96acb4a50e0fb4cbee69096d4f6285df33c7369e. Academia: https://www.academia.edu/2756208/Preferred_swing_ratio_in_jazz_as_a_function_of_tempo. ↩︎
Microtiming Deviations and Swing Feel in Jazz (2019). Scientific Reports. https://www.nature.com/articles/s41598-019-55981-3. The cross-genre comparison of microtiming magnitude (jazz vs rock) is one of the clearer empirical findings in the contemporary rhythm-perception literature. ↩︎
A swingogram representation for tracking micro-rhythmic variation in jazz performances (2017). Journal of New Music Research. https://www.tandfonline.com/doi/full/10.1080/09298215.2017.1367405. Broader review of microtiming-based descriptors of swing across performers and styles. ↩︎