A 2026 StarCraft II tournament study found competitors' wrist tremor declined over the day and was unrelated to game outcome or APM

In "Esports and Physiological Tremor: a StarCraft 2 Tournament Study" (Białecki et al., arXiv:2607.06577, 2026), wrist-accelerometer recordings from 16 male StarCraft II players at a live tournament deviated sharply from population norms — elevated tremor power in the 2–4 Hz band, reduced power at higher frequencies, with effect sizes of d = 1.6–2.3 against reference data. The counterintuitive finding is directional: rather than worsening with accumulated fatigue across the tournament day, the measured tremor declined, and it tracked neither match outcome nor actions-per-minute (APM).

The authors frame physiological tremor as "a sensitive neuromuscular indicator that may be modulated by cognitive load and competitive stress," and read the day-long decline as "a generalised psychophysiological adaptation to competitive esports rather than being a short-term performance predictor." In other words, the signal is interpreted as a trait-like feature of how a competitor's neuromuscular system settles under sustained competition, not a real-time gauge of who is about to win.

The study is one output of a small Polish research program that treats esports competitors as a distinct physiological population (see claim-elite-sc2-players-worse-grip-strength-than-reference-group). Methodologically it belongs to the same wearable-accelerometer sensing family as broader motion-foundation-model work (see claim-inertia-1-pretrained-18-million-hours-accelerometer), though at far smaller scale. Its clinical framing of extreme-repetition hand use echoes a much older occupational-medicine precedent (see claim-telegraphists-cramp-1890s-classified-as-occupational-neurosis).