June 30, 2026
Many athletes believe that more caffeine before a workout leads to better performance. But a new systematic review and meta-analysis suggests the relationship between caffeine and endurance is more complex. The study indicates that most of the benefit may come from lower doses, well before the amounts many athletes typically use.
About the study
Caffeine has been studied as a performance aid for decades. The evidence supporting its effects is well known. What is less clear is whether higher doses provide more benefit or simply more side effects. Researchers analyzed 48 randomized, placebo-controlled trials involving 689 participants. They compared how low, moderate, and high caffeine doses affected aerobic time-trial performance.
Low doses were defined as up to 3 milligrams per kilogram of body weight. Moderate doses ranged from 4 to 6 mg/kg. High doses were anything above 6 mg/kg. The team only included studies using pure caffeine in capsule or liquid form. They excluded coffee, energy drinks, and gum because caffeine content in those products can vary by more than 50 percent depending on how they are made.
Low doses nearly matched moderate doses
Low caffeine doses produced a statistically significant improvement in time-trial completion time. Moderate doses also improved performance significantly. But when researchers looked at actual race-time improvements, the real-world difference between the two dose ranges was almost negligible. Low doses corresponded to a mean improvement of about 2.14 percent. Moderate doses came in at about 2.18 percent.
Doubling or tripling caffeine intake may not get athletes meaningfully further across the finish line. The moderate-dose group showed more variability across studies, meaning the results were less consistent. The low-dose group showed a remarkably uniform effect, suggesting a more reliable performance boost. Researchers noted that two studies had an outsized influence on the moderate-dose results. Removing them from the analysis shifted the overall picture. No studies using high caffeine doses met the inclusion criteria for time-trial performance, so that end of the spectrum remains largely uncharacterized.
Why caffeine affects people differently
The review highlights the role genetics plays in how the body responds to caffeine. A gene called CYP1A2 controls an enzyme responsible for breaking down roughly 95 percent of the caffeine a person consumes. Depending on which version of this gene a person carries, caffeine may work very differently.
People who carry the CC variant of CYP1A2, commonly classified as slower caffeine metabolizers, may see a weaker or less consistent performance response, regardless of how much they take. Those who carry the AA or AC variants tend to show energetic benefits from caffeine more frequently. The review also points to early evidence that other genetic factors, including variations in adenosine receptors, may also play a role. That research is still developing, but it adds to the picture of why two people can take the same dose and have completely different experiences.
Finding the right pre-workout dose
For those who use caffeine before workouts or races, a straightforward starting point is that less may be enough. A low dose of roughly 1.3 to 3 mg/kg of body weight, taken about 60 minutes before exercise, produced significant and consistent performance improvements across the included studies. For a 70 kg person, that is about 90 to 210 mg of caffeine.
A standard cup of coffee contains anywhere from 80 to 200 mg of caffeine depending on how it is brewed. This means a morning cup may already put a person in that effective range. Most of the studies used caffeine capsules, which deliver a precise dose, but the numbers are a useful reference point.
The case for starting low is also reinforced by the side effects that come with higher doses. High caffeine intake is commonly linked to anxiety, heart palpitations, headaches, poor sleep, and gastrointestinal issues. If moderate doses offer only a marginal real-world advantage over low doses, the risk-benefit math shifts in favor of the smaller amount, especially for people sensitive to caffeine or training in the evening.
The takeaway
Low caffeine doses improved endurance time-trial performance by roughly 2 percent, nearly matching the real-world benefit of moderate doses. Because higher doses come with greater side-effect risk and no proportional performance gain, starting low is a sensible strategy for most people.
