The menstrual cycle and training
Evidence: contested
The claim that performance swings reliably across the menstrual cycle is not well supported. The best meta-analysis finds at most a trivial average effect, drawn from low-quality studies with high variation between people. Track symptoms and individualise; do not prescribe training by cycle phase from a chart.
Not medical advice
This is a general knowledge base, not medical advice. Hormonal contraception, menstrual disturbance and related decisions are medical and personal; discuss them with a doctor who knows your situation.
The idea that women should plan training around menstrual cycle phases has become popular, often promoted through apps and coaching products. The evidence does not support prescribing by phase. What the data show is a small, uncertain average effect sitting on top of large differences between individuals.
What the best evidence says
The most thorough synthesis pooled 78 studies. Performance was, on average, trivially lower in the early follicular phase, around the period itself, than in other phases. The effect was small, heterogeneity between studies was high, and study quality was poor: only 8% of studies were rated high quality and most were low or very low (McNulty et al. 2020). A trivial average effect drawn from weak studies is not a basis for confident scheduling.
A major reason the literature is so noisy is method. Most studies did not confirm which cycle phase a participant was actually in using hormone measurement, so ‘phases’ are frequently mislabelled, which both inflates and confuses the reported effects (Elliott-Sale et al. 2021).
Hormonal contraception
For runners on the combined pill, the picture is similar. A meta-analysis of oral contraceptives and performance found a pooled effect very close to zero, so they are unlikely to help or hinder performance on average (Elliott-Sale et al. 2020). Individual responses vary and the primary studies are again mixed in quality. Contraception choice is a medical and personal decision, not a performance lever.
Symptoms are real even when the average effect is not
A trivial average is not the same as 'no effect for you'
Some women experience genuine, disruptive symptoms, heavy bleeding, pain, fatigue, that clearly affect specific sessions. A near-zero group average can hide large individual experiences in both directions. The point is not that the cycle never matters, but that no fixed phase-based template captures who it matters for.
This is the individual variation problem in sharp form. The group mean describes an average woman who does not exist, while real symptoms vary enormously between people and between cycles.
What to do instead
Track the cycle alongside training and symptoms, then individualise from the pattern you actually see. If a runner reliably feels flat or sore at a certain point, adjust hard sessions around it for her, not because a chart says so. Treat phase-based prescriptions sold as universal rules with scepticism: the evidence under them is thin, and honest accounts of the field say so plainly (McNulty et al. 2020).
The cycle does become a clear warning sign in one direction. Missing periods is not a phase to train around but a red flag for low energy availability and RED-S, and it warrants attention rather than scheduling.