Reviews/Caffeine and Performance/Caffeine mouth rinse aids performance
Dosing · CAFF-DOS-03 · v2
Swilling a caffeine solution around the mouth and spitting it out, without swallowing it, seems to help with some kinds of exercise but not others. Several small, well-controlled trials found gains: people cycled longer before quitting [5], covered more ground in a running test [6], produced more sprint power [8], did more reps to failure with a high dose [2], and shot more accurately in a precision sport [13]. But the same body of work shows clear misses: no effect on one-rep maximum strength [2], no effect on fast knee movements [11], and little benefit during interval cycling after a meal [7]. Because the wins depend on the dose and the type of task, and every study was small, the panel treated the broad statement as supported in direction but not firmly settled.
Eight included crossover trials point mostly in favour but unevenly. Time to exhaustion on a bike rose 17% with the rinse [5], a 3-km cycling trial improved in a rinse arm [3], Yo-Yo intermittent-running distance increased [6], repeated-sprint mean and peak power rose [8], a high dose lifted muscular endurance reps to failure [2], and shooting accuracy improved [13]. Against these, the same corpus records nulls: no bench-press 1-RM change [2], no peak-torque effect at 180°/s [11], and no significant performance or physiological effect during post-meal interval exercise, only a difference in perceived exertion [7]. The positive results span different labs, populations and modalities, but samples are small (N=8 in [11], N=12–14 in several). The contested point: The division is whether a signal that appears only under certain doses, tasks and conditions supports an unqualified across-the-board statement. The Synthesist reads the positive core across modalities as real; the Sceptic reads the nulls on strength [2], high-velocity torque [11] and fed-state interval exercise [7] as evidence the effect is contingent rather than general.
All 13 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.
All included trials are small crossover studies (N=8 in [11], N=12-14 in several), where single-block or single-sprint positive findings are prone to inflated effect sizes and publication bias.
The unqualified claim overreaches: [2] found no 1-RM strength effect, [11] no effect at high angular velocity, and [7] only a marginal RPE difference with 'limited ergogenic benefit'. The effect is dose- and task-contingent, not general.
Where it remains uncertain. The same corpus records nulls on maximal strength [2], high-velocity torque [11] and post-meal interval exercise [7], and all trials are small.
Aleatory
Trials run from N=8 [11] to N=38 [3], so single-block or single-sprint effects carry wide sampling noise.
Epistemic
Only high-dose rinse moved muscular endurance while strength was unchanged in the same trial [2]; the dose threshold for a benefit is not pinned down.
Inferential
[3] used magnitude-based inference and the rinse advantage was restricted to early-of-day testing, weakening a clean performance inference.
The load-bearing papers from a corpus of 13, each classified by what it does to the claim. Every DOI is taken from the corpus the claim was graded on.
Dose-dependent muscular-endurance benefit; no 1-RM strength effect
“HCMR significantly increased muscular endurance performance ( p = 0.01) and decreased RPE values ( p = 0.01).”
Panel auditInternal validity: medium · scope of inference: Bench-press strength and muscular endurance in trained young males.Key limitations: n=14 · strength endpoint null · males only
Cycling time-trial benefit, conditional on time of day
“Rinse and Ingestion+Rinse elicited better performance relative to Placebo among subjects that performed prior to 10:00 h (Early) compared with after 10:00 h (Late).”
Panel auditInternal validity: medium · scope of inference: 3-km cycling time trials in recreational cyclists, stratified by CYP1A2 genotype and time of day.Key limitations: magnitude-based inference · benefit only in early subgroup · no p-values
Positive endurance performance effect
“Time until exhaustion increased 17% (effect size = 0.70) in CAF compared to PLA (p = 0.04).”
Panel auditInternal validity: medium · scope of inference: Cycling time-to-exhaustion at 80% respiratory compensation in twelve active men.Key limitations: n=12 · males only · RPE higher with caffeine
3 of 13 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.
A real but inconsistent, dose- and modality-contingent ergogenic signal from small trials supports the claim directionally without establishing it generally.
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.
Adequately powered trials. Every included trial is small (N=8 in [11], N=12–14 in several), leaving effect sizes imprecise and vulnerable to inflation. The panel rates this the highest-importance open gap for the claim, and names the study design that would close it: Pre-registered, adequately powered crossover trials on a fixed performance endpoint with declared minimal important difference.
Open question: Across which exercise modalities does caffeine mouth rinse reliably improve performance? Why it is still open: Positives in endurance [5], intermittent running [6] and sprint power [8] sit beside nulls in maximal strength [2] and high-velocity torque [11], with no cross-task synthesis.
Priors is a sceptical first reader, not peer review, and not medical advice.
Release 2026-q4 · graded 21 Oct 2026 · protocol v0.19