Does caffeine make you faster?

The evidence

Caffeine is one of the best-studied aids in endurance sport. A meta-analysis of 46 randomized controlled trials found that caffeine made endurance time trials about 2.2% faster (Southward and colleagues, 2018). The doses that worked were in the range of 3–6 mg per kilo of body weight.

For a 4-hour marathoner, 2.2% would be about 5 minutes. That figure is a rough illustration, though: the trials used time trials, not marathons, so the exact gain in a marathon is not known.

Caffeine in the only marathon fuelling trial

The only randomized trial of a fuelling plan in recreational marathoners included caffeine (Hansen and colleagues, 2014). The planned group took about 60 g of carbohydrate an hour, water, and about 90 mg of caffeine an hour. They finished about 11 minutes faster than runners who fuelled as they liked. Because the plan combined all three, the trial can’t say how much of the gain came from caffeine. See how many carbs per hour for the carbohydrate side.

How much and when

There are two common ways to take it. Both come from current sports-nutrition guidance (Guest and colleagues, 2021) and the dose range in the trials:

1. One dose before the start. About 3 mg per kilo, 45–60 minutes before the gun. The useful range is 2–6 mg per kilo.
2. A split dose. About 1–2 mg per kilo before the start, then about 1–1.5 mg per kilo in the final third of the race.

Whichever you choose, keep the total at or below 6 mg per kilo.

Worked example: a 70 kg runner

This is an example only. Multiply your own weight by the same numbers.

Plan	Per kilo	For 70 kg
Single dose before the start	3 mg/kg	210 mg
Full single-dose range	2–6 mg/kg	140–420 mg
Split: before the start	1–2 mg/kg	70–140 mg
Split: final third of the race	1–1.5 mg/kg	70–105 mg
Upper limit for the whole race	6 mg/kg	420 mg

The split plan adds up to 140–245 mg for a 70 kg runner, which is 2–3.5 mg per kilo.

For comparison, the Copenhagen plan’s 90 mg an hour would add up to 360 mg over a 4-hour race. For a 70 kg runner, that is about 5.1 mg per kilo, inside the 6 mg limit but near the top of the range.

Test it in training

Caffeine should be tested in training before race day. Use it in a long run or a tune-up race with the same timing and products you plan to use. Our suggestion: rehearse it alongside your carbohydrate plan, and check the label of every gel and drink you plan to take, since caffeine in fuel products adds to the total.

If you have a health condition or take regular medication, ask a doctor before using caffeine supplements. This page is general information.

What about other supplements?

Caffeine stands out because most other popular aids don’t hold up for marathon running:

• Sodium bicarbonate made a negligible difference in continuous running and caused stomach problems in 29.5% of users (Miller and colleagues, 2025).
• Nitrate has a small effect that disappears in women-only studies and in very fit athletes (Senefeld and colleagues, 2020).
• Ketone esters do not help endurance performance (Valenzuela and colleagues, 2020). See keto and fat adaptation.

What we don’t know

• The 2.2% comes from time trials, not marathons. No trial in our evidence base tested caffeine alone in a marathon.
• The split-dose timing is guidance, not something directly compared against a single dose in marathoners.
• Individual responses vary, which is why testing in training matters.
• The Copenhagen trial bundled caffeine with carbohydrate and water, so it can’t isolate caffeine’s share of the gain.

Sources

• Southward et al. (2018). Caffeine and endurance performance: meta-analysis of 46 trials.
• Guest et al. (2021). Caffeine and exercise performance: dosing guidance.
• Hansen et al. (2014). Planned vs self-chosen fuelling in the Copenhagen Marathon.
• Miller et al. (2025). Sodium bicarbonate and endurance performance.
• Senefeld et al. (2020). Nitrate supplementation and endurance performance.
• Valenzuela et al. (2020). Ketone supplements and endurance performance.