Caffeine Is the One Supplement Your Clients Are Right About. Here's Exactly How to Coach It.

How we sourced this. Every claim below is tied to a named study — an umbrella review, a meta-analysis, a position stand, a randomised trial, or the live trial registry — and linked. We also graded the evidence: where it’s strong, we say so; where it’s early or contested, we say that too. This piece was reviewed for accuracy before publishing and carries a real byline, because on health topics a citation you can’t stand behind is worse than no citation at all. Last updated 9 October 2026.

The supplement that actually earns the hype

Most of the supplements your clients ask about are oversold. Caffeine is the exception — the one they’re basically right about. Where magnesium’s marketing races miles ahead of the evidence, caffeine’s is, for once, roughly as good as advertised.

An umbrella review pooled 21 separate meta-analyses and found caffeine reliably improved aerobic endurance, muscle strength, muscle endurance, power, jumping and speed — backed by moderate-quality evidence.¹ The sport’s own governing body of nutrition agrees: the International Society of Sports Nutrition’s official position is that caffeine acutely enhances a broad range of endurance and anaerobic performance, in trained and untrained people alike.²

So this isn’t a “maybe.” The useful questions are how much, for whom, at what dose — and where it quietly bites back. That’s what a coach needs, and that’s what the rest of this is.

Endurance: the strongest, most bankable effect

If you coach anything with a clock on it, this is the headline. A meta-analysis of 46 trials found caffeine improved endurance time-trial performance by a small but consistent margin — mean power up about 3% and completion time down about 2% versus placebo — at moderate doses of 3–6 mg/kg.³ Two to four percent sounds trivial right up until you remember endurance events are routinely won by less than that.

The umbrella review makes the same point from the top down: caffeine’s effect is generally larger for aerobic than anaerobic work.¹ So if a client is training for a race, a long ride, a rowing test — anything sustained — this is where the few percent is most worth having.

Evidence grade: strong. This is the most consistent, best-replicated benefit caffeine has.

Strength and power: real, but don’t oversell it

For lifting, the effect is genuine but smaller. A meta-analysis of nine separate meta-analyses landed on a significant-but-modest boost — a standardised effect of about 0.18 for maximal strength and 0.30 for muscular endurance.⁴ In plain terms: your client will likely grind out a rep or two more and move the bar a touch faster — not add 10 kg to their squat overnight.

And the effect isn’t uniform. It shows up more at high loads (≥75% of 1RM) and in big lower-body lifts like the squat than in the bench press, which tracks with the larger muscle mass doing the work.⁷ Useful nuance if you’re deciding which session to time it around.

Evidence grade: moderate for strength and power — real, repeatable, but smaller than the endurance effect and load-dependent.

It works for women too — which the old research ignored

For years, “caffeine works” quietly meant “caffeine works for men,” because most studies were run on young males. A meta-analysis built specifically around women closed that gap: caffeine produced a significant ergogenic effect on both muscular endurance (SMD 0.25) and strength (0.18), similar in magnitude to what’s seen in men.⁵

That matters for how you coach a mixed roster — you don’t need to caveat caffeine differently for your female clients on the performance side. The one honest asterisk is the same as for men: lower-body-specific effects are less certain than upper-body ones.⁵

Evidence grade: moderate, and a welcome correction to a male-only evidence base.

Dose and timing: the part clients get wrong

Here’s where a coach earns their keep, because most people either under-do it or wildly over-do it. The evidence-based window is 3–6 mg/kg of body mass, taken about 60 minutes before training.² For a 70 kg client that’s roughly 210–420 mg — not a pre-workout scoop stacked on top of two coffees.

Two details worth passing on. Minimal effective doses may be as low as ~2 mg/kg, so smaller is often enough — in a five-condition trial in resistance-trained men, 2 mg/kg was the only dose that improved lower-body strength, all three doses added lower-body reps at 60% of 1RM, and upper-body strength rose modestly with dose.¹² The point isn’t that 2 beats 6; it’s that the floor is low and most of the benefit is already there at the smallest dose, so there’s little to gain by pushing. And pushing up to 9 mg/kg doesn’t buy more performance — it just buys more jitteriness, gut upset and wrecked sleep.²⁶ Source changes the timing, too: capsules want the full ~60 minutes, but caffeinated gum or coffee can work within 10–15 minutes.⁶

For the daily ceiling rather than the pre-workout dose, the most thorough safety review — 381 studies across five outcome areas — supports up to 400 mg/day in healthy adults as not associated with adverse cardiovascular, behavioural, bone or reproductive effects, 300 mg/day in pregnancy, and about 2.5 mg/kg/day for adolescents.¹⁵ A 6 mg/kg pre-workout in an 80-kg client is 480 mg on its own — already over the daily line before the morning coffee. That’s the arithmetic to show a client who stacks.

Evidence grade: strong on the dose range and the daily ceiling; this is settled enough to coach with confidence.

Individual response: the contested frontier

This is where an honest coach separates from a supplement rep. It’s fashionable to “personalise” caffeine by the CYP1A2 gene, which governs how fast you metabolise it. A 2026 randomised trial found the ergogenic effect was largest in fast-metaboliser AA carriers and minimal in CC carriers — and yet the same trial found genotype didn’t change the overall effect, and nobody turned “ergolytic,” i.e. worse than placebo.⁹

The trial that started the gene story points the other way, and you should know it. In 101 competitive athletes doing a 10-km cycling time trial, 4 mg/kg caffeine made fast-metaboliser AA carriers 6.8% faster, did nothing for AC carriers, and made slow-metaboliser CC carriers 13.7% slower than placebo — a genuine ergolytic effect in about one in ten athletes.¹⁰ So the two largest genotype trials disagree on the one thing that matters to a coach: whether anyone gets worse. Reviews of inter-individual variation add that genotype is only one driver — environment, prior caffeine exposure and the epigenetic legacy of it all shape the response, which is why a gene result alone is a weak predictor.¹³

One piece of this is settled, and it’s the one clients ask about most: habitual coffee drinking does not blunt the effect. When 40 trained cyclists were split by daily intake (about 60, 140 and 350 mg/day) and given 6 mg/kg, all three groups improved their time trial by the same margin.¹¹ Your three-espresso client doesn’t need to “wash out” before race day.

Pull back and it gets muddier still: other reviews conclude that CYP1A2 genotype doesn’t meaningfully modulate the average response at all.⁶ Translation for the gym floor: individual variation is real — some clients feel wired, some barely notice — but the gene test being sold as the oracle isn’t backed up yet.

Evidence grade: contested on genotype, strong that habitual use doesn’t matter. Coach it as an individual experiment (“let’s see how you respond”), never as a DNA readout — and start new clients at 2–3 mg/kg, so if they’re one of the rare poor responders they find out on a training day, not a race day.

The honest downsides

Caffeine’s real risks aren’t exotic — they’re sleep and blood pressure. It has a long half-life, so an afternoon dose can quietly dismantle the recovery you’re trying to build that night. The sleep cost now has numbers: across 24 studies, caffeine cut total sleep by 45 minutes and sleep efficiency by 7%, added 9 minutes to falling asleep and 12 minutes of waking in the night, and trimmed deep sleep — and the modelled cut-offs were about 8.8 hours before bed for a coffee and 13.2 hours for a ~220-mg pre-workout.¹⁴ A 6 pm training session with a scoop of pre-workout is, by that arithmetic, a midnight problem. And it reliably raises systolic blood pressure in the short term, which means genuine caution for anyone with hypertension or heart concerns — that’s a doctor conversation, not a pre-workout tip.⁸

Anxiety, jitteriness and gut upset also climb with the dose, which is one more reason the megadose approach backfires.⁶ The performance ceiling is low and the side-effect floor drops out fast.

Evidence grade: well-established. These aren’t fringe worries — they’re the predictable cost of getting the dose or the timing wrong.

What’s being tested next

The live registry shows where the field is heading — and it’s telling. Recruiting and about-to-start trials are testing caffeine dosing against CYP1A2 and ADORA2A genotype in cycling endurance, taurine-and-caffeine blends in professional footballers, and caffeine’s effect on fat oxidation during exercise.¹⁶

Notice the shift: almost nobody is still asking “does it work” — that’s settled. The open questions are for whom, in what combination, and can we actually personalise it. We’ll update this piece as those trials read out — that’s the point of an evidence page that’s alive instead of frozen.

Your script for Monday

When the next client asks, here’s the whole thing in plain language:

“Caffeine’s the rare supplement that mostly lives up to the hype. If you’ve got an endurance session or a hard lifting day, about 3 to 6 milligrams per kilo — for most people that’s one to three strong coffees — an hour before helps a real, if small, amount. More isn’t better: megadosing just makes you jittery and wrecks your sleep. Keep it out of the last nine hours before bed if you actually want to recover, stay under about 400 milligrams for the whole day, and don’t worry about ‘tolerance’ from your daily coffee, it doesn’t blunt the effect. If you’ve got blood-pressure or heart stuff going on, check with your doctor first. And ignore the ‘get your caffeine gene tested’ hype for now — we’ll just see how you respond.”

That’s it. No overpromising, no bro-science, no megadosing. Just the evidence, translated into something a human can act on — which is the entire job.

Sqwod reads the research so your clients don’t have to. This is coaching information, not medical advice; anything involving a medical condition, pregnancy, or medication belongs with a qualified professional. Spotted something that needs sharpening? That’s how an evidence page stays honest — tell us and we’ll correct it in public.

Caffeine affects sleep, anxiety and blood pressure: if you’re managing a diagnosed heart, sleep, or anxiety condition, loop in a professional rather than self-experimenting.