Caffeine dose response splits by sprint phase in trained 100-m runners
Acute caffeine did not behave like a uniform performance booster in trained male 100-m sprinters. In a randomized, double-blind, placebo-controlled crossover trial, 3 mg/kg and 6 mg/kg both improved several sprint-related outcomes, but the larger dose did not simply outperform the smaller one across every measure. The benefits clustered where sprint races are won or lost first: reaction, early acceleration, and the ability to produce force quickly off the blocks.
Sixty minutes after ingestion, 21 male 100-m sprinters completed placebo, 3 mg/kg caffeine, and 6 mg/kg caffeine conditions separated by at least 7 days. Each participant then performed two maximal 60-m sprints, followed by countermovement jump, squat jump, and a 10-s Wingate test. Investigators tracked cumulative 10-, 30-, and 60-m times, start reaction time, starting-block rate of force development, stride variables, and perceived exertion. The design matters because it tests the dose-response question in the setting most relevant to specialist sprinters rather than in a mixed athletic population or a repeated-sprint protocol.
The primary finding was straightforward. Both caffeine doses beat placebo for cumulative 10-, 30-, and 60-m times, with all treatment effects reaching statistical significance. Relative to placebo, 3 mg/kg shortened cumulative times by 0.056 s at 10 m, 0.083 s at 30 m, and 0.123 s at 60 m. The 6 mg/kg dose produced larger improvements of 0.075 s, 0.115 s, and 0.175 s, respectively. Even in sprinting, where fractions of a second decide placings, those differences are not trivial.
One segment resisted a simple dose escalation story. At 10 m, the additional 0.019-second reduction with 6 mg/kg over 3 mg/kg was not statistically significant in the primary mixed-effects model. By 30 m and 60 m, however, the higher dose did deliver further gains, with reductions of 0.032 s and 0.052 s compared with 3 mg/kg. In practical terms, the larger caffeine dose appeared to matter more as the sprint unfolded through the acceleration phase than at the very first steps out of the blocks.
That pattern fits the biology of caffeine as an adenosine-receptor antagonist. The compound does not directly make muscle stronger in a simple mechanical sense; instead, it reduces inhibitory signaling in the central nervous system, raises alertness, and can lower perceived effort. In a short sprint, those effects must be translated into sharper neural drive, faster force production, and more effective application of propulsive force. The data suggest that the translation is not identical across every phase of the race. Early reaction may respond to central stimulation alone, while sustained acceleration may depend more on how much additional force the athlete can generate and apply in the block exit and first strides.
Both doses also improved non-running outcomes relative to placebo. Reaction time improved, perceived exertion fell, jump height rose, Wingate peak and mean power increased, and starting-block rate of force development improved. Those changes strengthen the argument that the ergogenic effect was not confined to race timing alone. The athletes were not merely running faster by chance; they were expressing more explosive neuromuscular output across several tests that depend on rapid force generation.
The higher dose did not extend every benefit. Additional advantages with 6 mg/kg were seen for jump height and starting-block rate of force development, but not for reaction time, perceived exertion, or Wingate mean power. Wingate peak power also failed to improve further with the larger dose. Stride length and stride frequency stayed unchanged, which is a useful clue. The sprint improvement likely came from force-related qualities rather than a wholesale remodeling of running mechanics. The athletes did not take more steps or longer steps; they produced each step with more effective power in the early phase.
That finding narrows the bedside analogue in sports science: if a coach wants an acute caffeine strategy for a trained sprinter, more is not automatically better. A 3 mg/kg dose already improved multiple outcomes, and 6 mg/kg added selective gains rather than universal superiority. The larger dose may suit athletes who need maximal assistance in block power and the first 30 to 60 m, but the absence of consistent extra benefit across all endpoints argues against assuming a linear dose-response relationship.
The trial also helps separate single-race sprinting from repeated-sprint or team-sport protocols. Much of the existing caffeine literature draws from soccer, basketball, or rugby players, where short efforts are embedded in fatigue, recovery intervals, and mixed metabolic demands. A 100-m sprint is cleaner physiologically. It asks the athlete to accelerate, near-maximize velocity, and hold technique under severe time pressure. Under those conditions, caffeine’s benefit seems phase-specific rather than uniformly distributed across the race.
The study has limits that matter for interpretation. The sample was small and restricted to 21 male sprinters, so the findings cannot be assumed to apply to women, less-trained athletes, or other sprint distances. The comparison also focused on acute dosing, not chronic exposure or habitual caffeine use patterns. Even so, the trial offers one of the clearest direct comparisons of 3 mg/kg and 6 mg/kg in specialist 100-m runners. The key message is not that caffeine always helps, but that its performance effect is distributed unevenly: stronger in the start and acceleration phases, and more consistent for explosive force measures than for every running metric at once.