Reviews/Caffeine and Performance/Caffeine improves endurance
Endurance · CAFF-END-01 · v2
Borderline: within 4 points of the Established/Likely boundary, so a re-grade could land either side.
Across eight separate randomized, placebo-controlled trials in this corpus, swallowing caffeine before aerobic exercise consistently made people faster or able to keep going longer, in both cycling and running [1][3][4][6][9][11]. The gains ranged from a few percent to around fourteen percent on hard performance measures like time trials and time-to-exhaustion, rather than indirect markers. The effect held up in larger studies [1][6] as well as small ones, and across trained through to elite athletes. The panel also found real limits in the same corpus: one study shows the effect depends on a person's caffeine-metabolism gene and can even make certain carriers slower [1], another shows four weeks of daily caffeine wipes out the boost [12], and a third shows caffeine only undid a sleep-loss penalty rather than lifting a rested person above normal [13]. The panel treated these as conditions on who benefits and when, not as reasons to doubt the average effect.
Every aerobic-endurance trial carried shows caffeine improving a direct performance endpoint. In cycling, 4 mg·kg decreased 10-km time by 3% overall [1], time to exhaustion rose +14.4 ± 1.6% (P = 0.017) [3], and a ~30-min time trial improved significantly (P = 0.0002) in 40 cyclists [6]. Caffeine and coffee were ~5.0% faster than placebo in a 45-min time trial [9], and running time to exhaustion rose from 355 ± 41 to 375 ± 41 s (Δ19.4 ± 16.5 s; P < 0.001) in elite athletes [11]. Both 3 and 6 mg/kg doses improved exercise time in the heat [4]. The corpus also records that the acute benefit was no longer apparent after four weeks of chronic use [12]. The contested point: The division is over whether 'improves' can stand unqualified. [1] shows the direction and magnitude depend on CYP1A2 genotype, with 4 mg·kg increasing cycling time 13.7% (P = 0.04) in CC carriers, while [12] shows habituation abolishes the benefit and [13] shows caffeine only matched normal-sleep placebo (p = 0.96) rather than exceeding it.
All 14 papers behind this claim with their DOIs and what each does to it, the panel’s arbitration, the evidence gap map and the open research questions. Free, and sent to your inbox.
We’ll also send you other claims from Caffeine and Performance, one at a time, and tell you when a release or a pivotal trial changes a grade. Unsubscribe by replying.
One reviewer's only job is to attack the finding. The challenges below were weighed before the grade was set.
The claim is unqualified, but [1] shows caffeine was ergolytic in CYP1A2 CC carriers (4 mg/kg increased cycling time 13.7%, P=0.04) and had no effect in AC carriers; the benefit is genotype-dependent, not universal.
[12] shows four weeks of chronic low-dose caffeine abolished the acute performance benefit, so 'improves' holds for acute use in non-habituated or cycled users, not for chronic habitual consumers.
Where it remains uncertain. The effect is not universal: it reversed in CYP1A2 CC carriers (13.7% slower) [1], was abolished after four weeks of chronic intake [12], and only reversed a sleep-restriction deficit without exceeding rested placebo (p = 0.96) [13].
Aleatory
Response magnitude and direction vary by CYP1A2 genotype, with AA improving and CC worsening at 4 mg·kg [1]; individual-level improvement counts varied across tertiles [6].
Epistemic
Only [12] probes habituation over four weeks and finds tolerance; the long-term dose-response and cycling-off strategies are otherwise uncharacterised in this corpus.
The load-bearing papers from a corpus of 14, each classified by what it does to the claim. Every DOI is taken from the corpus the claim was graded on.
Largest trial (n=101); direct performance gain, but establishes the genotype-dependence caveat.
“Overall, 4 mg·kg caffeine decreased cycling time by 3% (mean ± SEM) versus placebo (17.6 ± 0.1 vs 18.1 ± 0.1 min, P = 0.01).”
Panel auditInternal validity: high · scope of inference: Competitive male athletes performing 10-km cycling time trials; effect modified by CYP1A2 genotype.Key limitations: male only · benefit genotype-dependent · performance worsened in CC carriers
Direct time-to-exhaustion improvement with mechanistic support (reduced central/peripheral fatigability).
“Compared with the placebo, caffeine increased time to exhaustion (+14.4 ± 1.6%, P = 0.017, 314.4 ± 47.9 vs 354.9 ± 40.8 s)”
Panel auditInternal validity: medium · scope of inference: Healthy males during high-intensity whole-body cycling to exhaustion.Key limitations: very small n=10 · male only · high-intensity cycling only
Dose-ranging improvement in endurance exercise time, including under heat stress.
“Both caffeine doses improved exercise time compared to placebo (3 mg/kg BW, p = 0.005; 6 mg/kg BW, p < 0.001).”
Panel auditInternal validity: medium · scope of inference: Participants performing incremental cycling endurance tests, emphasis on hot environment.Key limitations: small sample · incremental test to exhaustion · heat-focused design
Qualifies scope: caffeine restored a sleep-restriction deficit rather than exceeding rested baseline.
“Time to task failure was shorter (p < 0.001) in SRP (6.23 ± 2.11 min) than in HSP (7.68 ± 2.92 min) and SRC (7.83 ± 3.19 min), without differences between HSP and SRC (p = 0.96).”
Panel auditInternal validity: medium · scope of inference: Physically active men performing high-intensity exercise under sleep restriction; benefit is recovery of a deficit, not improvement over rested placebo.Key limitations: very small n=9 · no rested caffeine vs placebo comparison · high-intensity rather than classic endurance
4 of 14 papers shown. The full review lists all of them, with the panel’s arbitration →
Claims are graded at each quarterly release, and re-graded between releases only when a pivotal trial lands. Every change is a new version of the claim.
Uniform direction across multiple independent, blinded, placebo-controlled trials with hard performance endpoints carries an Established verdict, held just below near-certainty by genuine genotype and tolerance heterogeneity.
Release 2026-q4 re-graded every claim from its corpus under protocol v0.19. An earlier release's grade is shown for the record; it came from an earlier protocol and is not a movement.
Assumptions that carry weight but are not themselves established.
The panel records the gap that would most move the claim, and the study that would close it.
Female aerobic-endurance response. Carried endurance trials are overwhelmingly male ([1],[3],[9],[11]), so the mean effect's applicability to women in this domain is untested here. The panel rates this the highest-importance open gap for the claim, and names the study design that would close it: Adequately powered randomized placebo-controlled endurance time-trial trial in female athletes, stratified by menstrual/hormonal status.
Open question: Does acute caffeine improve aerobic endurance performance in women to the same degree observed in men? Why it is still open: The carried endurance trials are predominantly male-only ([1],[3],[9],[11]), leaving the female response unobserved in this corpus.
Priors is a sceptical first reader, not peer review, and not medical advice.
Release 2026-q4 · graded 21 Oct 2026 · protocol v0.19