Study analysis · Journal of the International Society of Sports Nutrition · 2024
Your morning coffee might be wasting your cycling potential — here's the exact dose that actually makes you faster.
If you're a cyclist, drinking 4 to 6 milligrams of caffeine per kilogram of your body weight makes you 3.2% faster, but less than that does nothing.
Overview
What the study found
The study in plain English — the bottom line, every takeaway we extracted, and what to do with them.
In simple terms
This study looked at lots of experiments where cyclists took caffeine or a fake pill and then raced. When you combine all those experiments, you can see that caffeine really does help them go faster — it’s not just luck. But it only works if they take the right amount — too little doesn’t help.
What’s the bottom line?
This study looked at whether drinking caffeine before a bike race helps cyclists finish faster.
How strong is this study?
This study is like a super careful science report that checked a bunch of good experiments and made sure they were fair — like making sure no one knew who got the real caffeine. That’s why we can trust the results more than if someone just asked a few friends. But even good studies can miss things, like if someone’s body breaks down caffeine differently.
40 / 100
- COI disclosure+40/40
- Data availabilitydata not shared
- Code availabilitycode not shared
19 / 100
- Randomizationnot randomized
- Blindingblinding unclear
- Control group+15/15
- Sample sizeno sample size reported
- Follow-upno follow-up reported
100 / 100
100 / 100
- P-values+15/15
- Effect size+20/20
- Confidence intervals+15/15
- Pre-registration+15/15
Each component is scored out of 100 and then capped by the study design — a case series cannot reach the ceiling a randomised trial can, however well it is reported.
Where it sits
RCT reviewsReviews of RCTs (Meta-analyses)
Max 100Randomized TrialsRandomized Trials
Max 90Reviews of Cohort StudiesReviews of Cohort Studies
Max 85Cohort StudiesCohort Studies
Max 72Reviews of Case-Control StudiesReviews of Case-Control Studies
Max 63Case-Control StudiesCase-Control Studies
Max 58Cross-Sectional & Case SeriesCross-Sectional & Case Series
Max 50Expert OpinionExpert Opinion
Max 560 / 100
Probability of being correct
Systematic reviews and meta-analyses of cohort studies. They sit above a single cohort study but below a single randomized trial, because the underlying evidence is still observational.
This design can establish causation. This study is a systematic review and meta-analysis of randomized controlled trials (RCTs), which are the gold standard for establishing causation. The included studies were crossover, placebo-controlled RCTs, and the meta-analysis combines their results to show a consistent effect of caffeine on cycling performance, supporting a causal relationship.
No Conflicts
No conflicts of interest identified
No conflicts of interest or funding sources were disclosed in the study. No industry ties or funder involvement were reported.
Independent Analysis Safeguards
- Study registered in PROSPERO (CRD42022358446)
- PRISMA guidelines followed
- Dual independent reviewers for study selection and data extraction
- Cochrane risk of bias assessment tool used
- Meta-analysis conducted using Review Manager 5.4 and Stata 14.0
- Publication bias assessed via funnel plots, Egger’s regression, and trim-and-fill tests
The study explicitly states that 'none of the studies reported commercial sponsorship,' and no funding or conflict of interest declarations were provided by the authors. The rigorous methodological safeguards (PROSPERO registration, dual review, bias assessment) support the independence and reliability of the analysis.
Key takeaways
- 01
Cyclists who took 4–6 mg of caffeine per kg of body weight finished about 3.2% faster and produced 4.3% more power.
- 02
Those who took 1–3 mg/kg saw no real improvement.
- 03
Yes — a 3% speed boost in a race could mean the difference between winning a medal and finishing 8th.
Surprising findings
- Publication bias was detected, but it didn’t change the conclusion.Most meta-analyses with publication bias are dismissed as unreliable — but here, even after statistical correction (trim-and-fill), the 3.2% performance boost remained significant.
- Caffeine’s effect on power and time were nearly identical — suggesting it doesn’t change pacing, just effort.You’d expect caffeine to make people go harder at the start and fade — but the 3.2% speed gain matched the 4.3% power gain, meaning cyclists maintained consistent, higher effort throughout.
Practical takeaways
Weigh yourself, then multiply by 4.5 to get your ideal caffeine dose in mg (e.g., 70kg × 4.5 = 315mg). Take it 60 minutes before your ride.
This dose is for trained cyclists — beginners or non-athletes may experience jitters or GI issues. Start lower and test in training, not race day.
high confidenceWhy this study matters
The Magic Dose: 4–6 mg/kg
Cyclists who took 4–6 mg of caffeine per kilogram of body weight improved their time-trial performance by 3.2% and increased power output by 4.3%. This was statistically significant (p < 0.01), while lower doses showed no meaningful benefit.
Most people think any coffee before a workout helps — but this study proves you need a precise, almost athletic-grade dose. Your latte won't cut it.
Low Dose? No Effect.
Doses of 1–3 mg/kg (about 70–210mg for a 70kg person) showed no significant improvement in completion time (SMD = -0.34, p = 0.19) or power output (SMD = 0.31, p = 0.07), despite being common in energy drinks and pre-workouts.
You’ve been told ‘any caffeine helps’ — but this meta-analysis of 15 studies says that’s a myth. You’re spending money and risking jitters for zero gain.
It Works for Everyone — Even If You're Not an Olympian
The benefits were consistent across age, VO2max levels, and race duration — whether you're a 25-year-old with high aerobic capacity or a 40-year-old with moderate fitness, the same dose worked.
Most performance hacks only work for elite athletes. This one works for weekend warriors too — if you get the dose right.
Want the whole report?
Detailed mode opens the full scientific breakdown — every score component, the methodology, conflicts of interest, the evidence analysis behind each claim, and the raw study data.
Overview
What the study found
The study in plain English — the bottom line, every takeaway we extracted, and what to do with them.
Not medical advice. For informational purposes only. Always consult a healthcare professional. Terms
This study looked at whether drinking caffeine before a bike race helps cyclists finish faster.
Research results
Cyclists who took 4–6 mg of caffeine per kg of body weight finished about 3.2% faster and produced 4.3% more power. Those who took 1–3 mg/kg saw no real improvement.
What this means - more context
Yes — a 3% speed boost in a race could mean the difference between winning a medal and finishing 8th.
This meta-analysis evaluates whether caffeine improves cycling time-trial performance and identifies the optimal dose.
Moderate caffeine (4–6 mg/kg) significantly improves cycling time-trial performance by reducing completion time by ~3.2% and increasing mean power output by ~4.3%, while low doses (1–3 mg/kg) show no significant effect. Benefits are consistent across age, VO2max, and exercise duration.
Methods Used
Systematic review and meta-analysis of 15 crossover, placebo-controlled studies in cyclists; used random-effects model to calculate standardized mean differences (SMD) for completion time and mean power output; assessed bias via Cochrane tool and publication bias via funnel plots, Egger’s test, and trim-and-fill.
Main Finding
Moderate caffeine (4–6 mg/kg) significantly improved cycling time-trial performance: SMD = -0.55 for completion time (p < 0.01) and SMD = 0.44 for mean power output (p < 0.05); low doses (1–3 mg/kg) showed non-significant effects (SMD Time = -0.34, p = 0.19; SMD MPO = 0.31, p = 0.07).
Confidence Level
High — pre-registered (PROSPERO), PRISMA-compliant, rigorous bias assessment, consistent effect sizes across studies, publication bias accounted for and did not alter conclusions.
Study Flags
Red Flags
- •Moderate risk of bias in 12 of 15 included studies
- •Potential publication bias detected (Egger’s test p < 0.05)
- •Limited subgroup analysis for age, genetics, and gender
Surprising Findings
Publication bias was detected, but it didn’t change the conclusion.
Most meta-analyses with publication bias are dismissed as unreliable — but here, even after statistical correction (trim-and-fill), the 3.2% performance boost remained significant.
Practical Takeaways
Weigh yourself, then multiply by 4.5 to get your ideal caffeine dose in mg (e.g., 70kg × 4.5 = 315mg). Take it 60 minutes before your ride.
RCT reviewsReviews of RCTs (Meta-analyses)
Max 100Randomized TrialsRandomized Trials
Max 90Reviews of Cohort StudiesReviews of Cohort Studies
Max 85Cohort StudiesCohort Studies
Max 72Reviews of Case-Control StudiesReviews of Case-Control Studies
Max 63Case-Control StudiesCase-Control Studies
Max 58Cross-Sectional & Case SeriesCross-Sectional & Case Series
Max 50Expert OpinionExpert Opinion
Max 560 / 100
Probability of being correct
Systematic reviews and meta-analyses of cohort studies. They sit above a single cohort study but below a single randomized trial, because the underlying evidence is still observational.
Human Meta-Analysis
Subject
Moderate probability
on the GRADE evidence scale
This study looked at lots of experiments where cyclists took caffeine or a fake pill and then raced. When you combine all those experiments, you can see that caffeine really does help them go faster — it’s not just luck. But it only works if they take the right amount — too little doesn’t help.
No conflicts of interest were detected in this study. No score impact.
Strengths
- Followed PRISMA guidelines and was prospectively registered
- Used comprehensive database searches with no language restrictions
- Included only high-quality crossover RCTs with placebo controls
Weaknesses
- Included studies varied in caffeine timing and delivery method (capsules vs. liquid)
- Some included studies had unclear or moderate risk of bias (12/15)
- Limited number of studies in low-dose subgroup (n=3 for Time, n=6 for MPO)
Methodology
Evidence Keywords
Statistical Reporting
Not medical advice. For informational purposes only. Always consult a healthcare professional. Terms
This study looked at whether drinking caffeine before a bike race helps cyclists finish faster.
Research results
Cyclists who took 4–6 mg of caffeine per kg of body weight finished about 3.2% faster and produced 4.3% more power. Those who took 1–3 mg/kg saw no real improvement.
What this means - more context
Yes — a 3% speed boost in a race could mean the difference between winning a medal and finishing 8th.
This meta-analysis evaluates whether caffeine improves cycling time-trial performance and identifies the optimal dose.
Moderate caffeine (4–6 mg/kg) significantly improves cycling time-trial performance by reducing completion time by ~3.2% and increasing mean power output by ~4.3%, while low doses (1–3 mg/kg) show no significant effect. Benefits are consistent across age, VO2max, and exercise duration.
Methods Used
Systematic review and meta-analysis of 15 crossover, placebo-controlled studies in cyclists; used random-effects model to calculate standardized mean differences (SMD) for completion time and mean power output; assessed bias via Cochrane tool and publication bias via funnel plots, Egger’s test, and trim-and-fill.
Main Finding
Moderate caffeine (4–6 mg/kg) significantly improved cycling time-trial performance: SMD = -0.55 for completion time (p < 0.01) and SMD = 0.44 for mean power output (p < 0.05); low doses (1–3 mg/kg) showed non-significant effects (SMD Time = -0.34, p = 0.19; SMD MPO = 0.31, p = 0.07).
Confidence Level
High — pre-registered (PROSPERO), PRISMA-compliant, rigorous bias assessment, consistent effect sizes across studies, publication bias accounted for and did not alter conclusions.
Study Flags
Red Flags
- •Moderate risk of bias in 12 of 15 included studies
- •Potential publication bias detected (Egger’s test p < 0.05)
- •Limited subgroup analysis for age, genetics, and gender
Surprising Findings
Publication bias was detected, but it didn’t change the conclusion.
Most meta-analyses with publication bias are dismissed as unreliable — but here, even after statistical correction (trim-and-fill), the 3.2% performance boost remained significant.
Practical Takeaways
Weigh yourself, then multiply by 4.5 to get your ideal caffeine dose in mg (e.g., 70kg × 4.5 = 315mg). Take it 60 minutes before your ride.
RCT reviewsReviews of RCTs (Meta-analyses)
Max 100Randomized TrialsRandomized Trials
Max 90Reviews of Cohort StudiesReviews of Cohort Studies
Max 85Cohort StudiesCohort Studies
Max 72Reviews of Case-Control StudiesReviews of Case-Control Studies
Max 63Case-Control StudiesCase-Control Studies
Max 58Cross-Sectional & Case SeriesCross-Sectional & Case Series
Max 50Expert OpinionExpert Opinion
Max 560 / 100
Probability of being correct
Systematic reviews and meta-analyses of cohort studies. They sit above a single cohort study but below a single randomized trial, because the underlying evidence is still observational.
Human Meta-Analysis
Subject
Moderate probability
on the GRADE evidence scale
This study looked at lots of experiments where cyclists took caffeine or a fake pill and then raced. When you combine all those experiments, you can see that caffeine really does help them go faster — it’s not just luck. But it only works if they take the right amount — too little doesn’t help.
No conflicts of interest were detected in this study. No score impact.
Strengths
- Followed PRISMA guidelines and was prospectively registered
- Used comprehensive database searches with no language restrictions
- Included only high-quality crossover RCTs with placebo controls
Weaknesses
- Included studies varied in caffeine timing and delivery method (capsules vs. liquid)
- Some included studies had unclear or moderate risk of bias (12/15)
- Limited number of studies in low-dose subgroup (n=3 for Time, n=6 for MPO)
Methodology
Evidence Keywords
Statistical Reporting
Scoring
How strong is this study?
This study is like a super careful science report that checked a bunch of good experiments and made sure they were fair — like making sure no one knew who got the real caffeine. That’s why we can trust the results more than if someone just asked a few friends. But even good studies can miss things, like if someone’s body breaks down caffeine differently.
40 / 100
- COI disclosure+40/40
- Data availabilitydata not shared
- Code availabilitycode not shared
19 / 100
- Randomizationnot randomized
- Blindingblinding unclear
- Control group+15/15
- Sample sizeno sample size reported
- Follow-upno follow-up reported
100 / 100
100 / 100
- P-values+15/15
- Effect size+20/20
- Confidence intervals+15/15
- Pre-registration+15/15
Each component is scored out of 100 and then capped by the study design — a case series cannot reach the ceiling a randomised trial can, however well it is reported.
Where it sits
RCT reviewsReviews of RCTs (Meta-analyses)
Max 100Randomized TrialsRandomized Trials
Max 90Reviews of Cohort StudiesReviews of Cohort Studies
Max 85Cohort StudiesCohort Studies
Max 72Reviews of Case-Control StudiesReviews of Case-Control Studies
Max 63Case-Control StudiesCase-Control Studies
Max 58Cross-Sectional & Case SeriesCross-Sectional & Case Series
Max 50Expert OpinionExpert Opinion
Max 560 / 100
Probability of being correct
Systematic reviews and meta-analyses of cohort studies. They sit above a single cohort study but below a single randomized trial, because the underlying evidence is still observational.
This design can establish causation. This study is a systematic review and meta-analysis of randomized controlled trials (RCTs), which are the gold standard for establishing causation. The included studies were crossover, placebo-controlled RCTs, and the meta-analysis combines their results to show a consistent effect of caffeine on cycling performance, supporting a causal relationship.
No Conflicts
No conflicts of interest identified
No conflicts of interest or funding sources were disclosed in the study. No industry ties or funder involvement were reported.
Independent Analysis Safeguards
- Study registered in PROSPERO (CRD42022358446)
- PRISMA guidelines followed
- Dual independent reviewers for study selection and data extraction
- Cochrane risk of bias assessment tool used
- Meta-analysis conducted using Review Manager 5.4 and Stata 14.0
- Publication bias assessed via funnel plots, Egger’s regression, and trim-and-fill tests
The study explicitly states that 'none of the studies reported commercial sponsorship,' and no funding or conflict of interest declarations were provided by the authors. The rigorous methodological safeguards (PROSPERO registration, dual review, bias assessment) support the independence and reliability of the analysis.