Caffeine and running performance: what the research shows

Most performance supplements promise more than they deliver. Caffeine is the rare exception. It is one of the most studied compounds in sports science, it is cheap, it is legal, and across hundreds of trials it reliably makes hard endurance work feel easier and turn out faster. The honest question is not whether it works, but how to use it well.

This article walks through what the research actually shows: the effective dose, the timing, the mechanism behind the effect, whether coffee is as good as a tablet, how much individual variation matters, and why the fear of building tolerance is largely a myth.

How strong is the evidence, really?

Very strong, by the standards of sports nutrition. Grgic and colleagues (2020) pulled together every published meta analysis on caffeine and exercise into a single umbrella review, covering twenty one meta analyses. They concluded that caffeine is ergogenic for aerobic endurance, muscle strength, muscle endurance, power and exercise speed, with the effect generally larger for aerobic than for anaerobic work. When the whole literature points the same way, you can stop wondering whether the effect is real.

For runners specifically, Ganio and colleagues (2009) reviewed time trial studies of more than five minutes across cycling, running, rowing and other endurance modes. The average performance benefit was about 3 percent. That sounds small until you translate it: roughly a minute knocked off a good 10 km, or several minutes off a marathon. The International Society of Sports Nutrition position stand, led by Guest and colleagues (2021), reaches the same headline conclusion and notes that aerobic endurance is the form of exercise that benefits most consistently from caffeine.

The effective dose and when to take it

The dose that comes up again and again is 3 to 6 mg of caffeine per kilogram of body mass, taken about 60 minutes before exercise. For a 70 kg runner that is roughly 210 to 420 mg, which is around two to four cups of brewed coffee. The ISSN position stand settles on this 3 to 6 mg/kg range as the evidence based dose, and notes that larger amounts do not produce larger benefits.

The 60 minute window reflects how the body handles caffeine. Blood levels peak roughly 45 to 60 minutes after ingestion, so taking it about an hour before the start lines the peak up with the early, decisive part of your run. Timing your fuelling around the same window pairs naturally with your carbohydrate strategy for long runs and races, and many caffeinated gels conveniently deliver both at once.

Why it works: blunting perceived effort

Caffeine is not a stimulant in the simple sense of giving you raw extra energy. Its main mechanism is in the brain. Caffeine is structurally similar to adenosine and competes for the same receptors. Adenosine normally accumulates during exercise and signals fatigue, dampening arousal and raising your sense of effort. By blocking those receptors, caffeine keeps that fatigue signal quieter for longer.

The practical consequence is measurable. Doherty and Smith (2005) ran a meta analysis of studies of caffeine and rating of perceived exertion and found that caffeine reduced perceived effort during exercise by about 5.6 percent. In other words, a given pace simply feels easier on caffeine. That lower perceived effort accounts for a meaningful share of the performance gain, because a pace that feels more comfortable is a pace you can hold for longer.

Low doses work too

You do not need to load a large amount to benefit. Spriet (2014) reviewed the evidence on low doses, below about 3 mg/kg or roughly 100 to 200 mg, and found them ergogenic in many exercise situations while improving alertness, mood and vigilance with fewer of the unwanted effects of high doses. Importantly, low doses also work when taken partway through prolonged exercise, not only beforehand.

That has a useful practical implication. A small caffeine hit late in a long run or marathon, exactly when fatigue is biting hardest, can lift you for the closing miles without the early jitters of a big dose taken up front. If caffeine has made you anxious or queasy in the past, starting at the low end is the sensible move.

Individual variation and the CYP1A2 gene

Caffeine does not affect everyone equally, and some of that difference is genetic. The liver enzyme that clears caffeine is encoded by the CYP1A2 gene, and a common variant determines whether you are a fast or slow metaboliser. Guest and colleagues (2018) tested this directly in 101 competitive athletes over a 10 km cycling time trial. Caffeine improved performance, but the benefit was concentrated in those with the AA genotype, who metabolise caffeine quickly. In the CC group, who metabolise it slowly, the high 4 mg/kg dose actually made performance worse.

You do not need a gene test to act on this. The lesson is that response is individual, so treat your own training runs as the experiment. If a standard dose leaves you wired, anxious or slower rather than sharper, you may be a slower metaboliser who does better on a smaller amount, or none at all before shorter efforts.

The habituation myth, coffee, and gut considerations

A persistent worry is that daily coffee drinkers go numb to caffeine and must abstain for days to feel an effect on race day. The evidence does not support a strict version of this. Habitual users still gain an ergogenic benefit from a dose taken before exercise, so a long caffeine taper is not required for most runners. If anything, the bigger risk in a taper over several days is the headache and grogginess of withdrawal arriving on race week.

Coffee itself is a perfectly good delivery vehicle. Trials comparing coffee with anhydrous caffeine find similar endurance benefits when the dose is matched. The only real drawback is precision: the caffeine in a home or cafe brew varies a great deal, so for an important race a measured tablet or a labelled gel makes the dose easier to control. Coffee also brings a known risk of needing the toilet, which some runners want to avoid during a race.

One last point of context: caffeine sharpens an engine, it does not build one. The aerobic capacity it helps you express still comes from training, which is where work on your VO₂max and your aerobic base does the heavy lifting. Caffeine is best thought of as a small, reliable bonus stacked on top of fitness, much like other evidence based aids such as beetroot juice for runners.

Frequently asked questions

Related reading: beetroot juice for runners: does it actually improve performance.