Caffeine does help strength, but the effect is small. A 2018 meta-analysis of ten studies found a small improvement in one-rep max, clear for upper-body lifts and not clear for lower-body ones. The best-supported dose is 3 to 6 mg per kilo of body weight, taken about an hour before training.
That is a modest return, and it comes with a cost if you train in the evening. This guide covers what the reviews measured, how to work out a dose for your body weight, when to take it, what it can do to your sleep, and how to test it on your own training log instead of trusting a pre-workout label.
Does caffeine improve strength?
Yes, by a small amount. A 2018 meta-analysis found caffeine improved maximal strength and power compared with placebo. The gain was clearer for upper-body lifts such as the bench press than for lower-body lifts such as the squat. It is a minor edge, not a replacement for good training.
The review, by Grgic and colleagues (2018), pooled ten studies that measured one-rep max and ten that measured power through vertical jump. Strength improved by a standardised mean difference (SMD) of 0.20 (95% CI 0.03 to 0.36, p = 0.023). Power improved by 0.17 (95% CI 0.00 to 0.34, p = 0.047). An SMD around 0.2 is conventionally read as a small effect.
Upper-body strength improved significantly (SMD 0.21, 95% CI 0.02 to 0.39). Lower-body strength did not reach significance (SMD 0.15, 95% CI -0.05 to 0.34). That does not prove caffeine does nothing for your squat, because the lower-body confidence interval still includes a positive effect. It means the evidence is weaker there.
What do the reviews actually show?
Across several reviews, caffeine improves muscle strength, muscle endurance and power by small to moderate amounts, and the evidence quality is mostly moderate. The main limits are that most participants were young men and that blinding was often poorly controlled, so expectation may inflate some results.
| Source | Who and what | Finding |
|---|---|---|
| Grgic et al. (2018) | Meta-analysis, 10 strength studies and 10 power studies | Strength SMD 0.20 and power SMD 0.17; upper-body strength improved, lower-body did not reach significance |
| Grgic et al. (2020) | Umbrella review of 21 meta-analyses in 11 reviews | Caffeine was ergogenic for strength, muscle endurance, power, jumping, speed and aerobic endurance; moderate-quality evidence for strength and muscle endurance |
| Guest et al. (2021) | International Society of Sports Nutrition position stand | Small to moderate benefits for muscular endurance, movement velocity and strength; 3 to 6 mg/kg consistently improves performance |
The 2020 umbrella review added two cautions. Not every analysis gave a definite direction once its 95% prediction interval was considered, which means a new study could plausibly find a different result. It also found the effect was generally larger for aerobic exercise than for anaerobic exercise, so lifters should expect less than a distance runner would.
The same review noted that most individual studies were done in young men, and called for more research in women, middle-aged adults and older adults. If you are outside that group, treat the numbers as a starting point for your own testing.
How much caffeine per kilo of body weight?
The best-supported dose is 3 to 6 mg per kilo of body mass. The International Society of Sports Nutrition (ISSN) says that range consistently improves performance, that the minimum effective dose is unclear but may be as low as 2 mg/kg, and that 9 mg/kg causes more side effects without being needed.
Those figures come from the ISSN position stand on caffeine and exercise performance. To find your dose, multiply your body weight in kilos by the dose per kilo.
| Body weight | 2 mg/kg | 3 mg/kg | 6 mg/kg |
|---|---|---|---|
| 60 kg | 120 mg | 180 mg | 360 mg |
| 80 kg | 160 mg | 240 mg | 480 mg |
| 100 kg | 200 mg | 300 mg | 600 mg |
Caffeine content varies a lot between coffee, energy drinks and pre-workout powders, so read the label or the manufacturer's information instead of guessing from the serve size. A scoop of pre-workout may already contain more than you intend, and it may contain other ingredients that this article does not cover.
The ISSN also lists genetic differences in caffeine metabolism and habitual intake as reasons people respond differently. Starting near the low end of the range and moving up only if you see no benefit is a sensible approach. It is the author's reasoning rather than a tested protocol, but it follows from the evidence that higher doses mostly add side effects.
When should you take it?
The most common timing in the research is about 60 minutes before exercise. The ISSN says the best timing likely depends on the caffeine source. For example, caffeine chewing gum may need a shorter wait before training than capsules do.
That is the extent of what the position stand supports, so avoid precise claims beyond it. If you drink coffee, the 60-minute guide is a reasonable place to start. If you use gum or a pre-workout powder, check the product information and test your own timing.
The position stand also reports that caffeine improved performance in both trained and untrained people, and that energy drinks and caffeinated pre-workout supplements have been shown to improve anaerobic and aerobic performance. Those products work, but the caffeine in them is most likely doing the job, so work out the milligrams before you stack extra coffee on top.
If you are also considering creatine, it works differently: it is taken daily to raise muscle stores, not timed before a session. The creatine and strength guide covers the doses and the average gains from meta-analyses.
Timing matters most for the heaviest work. If you plan a top set or a PR attempt, take caffeine ahead of it. For lighter accessory days, you may find the benefit is not worth the sleep cost described below.
What does caffeine cost your sleep?
Caffeine can disturb sleep even when taken hours before bed. In a home study of 400 mg doses, caffeine taken at bedtime, 3 hours before bed and 6 hours before bed each significantly disturbed sleep compared with placebo. The authors supported avoiding substantial caffeine for at least 6 hours before bed.
The study, by Drake and colleagues (2013), used both self-reported sleep and a validated portable monitor. A 400 mg dose is at the high end of the ISSN range for many lifters, so a smaller dose may disturb sleep less, but that study did not test one.
That matters because sleep affects the same lifts you are trying to improve. As covered in sleep and strength training, a systematic review found repeated nights of restricted sleep reduced force output on multi-joint lifts such as the squat, bench press and deadlift. If a late session fuelled by caffeine costs you sleep for several nights, you can lose more than the supplement gives you.
The practical consequence is a trade-off for evening lifters:
- Train earlier if you can. Morning and lunchtime sessions leave more hours between the dose and bedtime.
- Lower the dose for late sessions. Staying near the low end of the range costs you less sleep than the top end, though the research gives no exact threshold.
- Skip it on non-critical days. Reserve caffeine for sessions where a heavy top set matters.
- Watch your sleep after any change. If your sleep worsens, adjust before you blame your programme.
The ISSN position stand also notes that caffeine may help performance in some people who are sleep deprived. That is a way to get through a bad day, not a strategy for training under repeated short sleep.
How do you test caffeine in your own training log?
Compare effort, not just load. Pick one lift, repeat it on comparable days at the same time, alternate caffeine and no caffeine, and record RPE for each set. If caffeine works for you, the same load and reps will feel easier, or the same effort will carry more load.
Group results are an average. The ISSN notes individual differences from genetics and habitual intake, so your response may be bigger or smaller than the reviews suggest. A short test beats a label claim.
A simple method:
- Choose one lift, such as the bench press, and keep the programme, rest times and time of day the same.
- Run two to four comparable sessions with your usual caffeine intake and two to four with a defined dose.
- Record load, reps and RPE (or RIR) for every working set.
- Compare RPE at matching load and reps, and look at your estimated one-rep max across the sessions.
- Note your sleep that night. A gain that costs you rest is not a clean win.
Habitual coffee drinkers should be careful reading the result. The ISSN lists habitual intake as one reason responses differ, so your normal daily coffee may blunt the effect of a pre-workout dose. This is not a reason to raise the dose past the research range.
You will also never fully escape expectation. The reviews flagged blinding as a weakness, which means knowing you took caffeine can change how a set feels. Judging by RPE alone is the weakest comparison, so lean on repeatable numbers such as bar load, reps and estimated one-rep max, and treat a single session as noise. The workout tracking metrics guide covers which fields are worth recording.
IronLedger logs load, reps and RPE in half steps or RIR for each set, with free-text notes on the set that stay in your history, as listed on the product page. It has no caffeine or supplement field, so you would type something like "caffeine 240 mg, 60 min before" into the note by hand. That is enough to filter your history later and compare matching sessions. For more on logging effort, see the RPE workout tracker guide.
What mistakes should you avoid?
Most mistakes come from treating a small, average effect as a guarantee: taking too much, ignoring the sleep cost, or crediting caffeine for progress that came from the training.
- Chasing a bigger dose. The ISSN says 9 mg/kg is associated with a high incidence of side effects and is not needed to get the benefit.
- Ignoring sleep. Caffeine 6 hours before bed still disturbed sleep in the home study, and poor sleep can hurt the lifts you are trying to improve.
- Using it to hide a strength plateau. If you have stalled, check programming, recovery and effort first. A small boost on the day does not fix a stalled programme.
- Judging from one session. Single sessions swing with sleep, food and stress. Compare several.
- Not counting the milligrams. Pre-workouts and energy drinks differ widely, and the label is the only reliable guide.
- Forgetting what else is in your evening. Alcohol also disrupts sleep, and the alcohol and strength training guide covers how it compares for recovery.
- Treating it as required. You can make strength progress without caffeine. If you have a heart condition, are pregnant or take medication, talk to a doctor before using it.
Frequently asked questions
Slightly. A 2018 meta-analysis found a small improvement in one-rep max (SMD 0.20), significant for upper-body strength but not lower-body strength. Treat it as a minor edge, not a major driver of progress.
Sources and references
- Grgic et al., Effects of caffeine intake on muscle strength and power: a systematic review and meta-analysis — caffeine improved strength (SMD 0.20) and power (SMD 0.17); upper-body strength improved, lower-body did not reach significance (2018).
- Grgic et al., Wake up and smell the coffee: caffeine supplementation and exercise performance, an umbrella review of 21 published meta-analyses — caffeine was ergogenic across strength, muscle endurance, power and aerobic endurance, with most studies in young men (2020).
- Guest et al., International Society of Sports Nutrition position stand: caffeine and exercise performance — 3 to 6 mg/kg consistently improves performance; about 60 minutes pre-exercise is the most common timing (2021).
- Drake et al., Caffeine effects on sleep taken 0, 3, or 6 hours before going to bed — 400 mg of caffeine at each time significantly disturbed sleep versus placebo (2013).
- IronLedger product page — per-set RPE in half steps, RIR and notes kept in history.
Keep learning
