Your Cycle and Exercise: What the Evidence Supports, and What It Does Not

The evidence does not support training by cycle phase, and the researchers who gathered it say so themselves. A meta-analysis of 78 studies found performance might be trivially reduced in the early follicular phase — a median pooled effect of −0.06, credible interval −0.16 to 0.04, on evidence classified as low quality — and concluded that general guidelines on exercise performance across the cycle cannot be formed. An umbrella review of strength research found that 9 of 10 moderate-to-high-quality studies saw no difference between phases, and called published recommendations to train by phase premature and potentially misleading. A third paper goes further and questions whether anybody outside a laboratory knows which phase a woman is in at all. None of that means the cycle does nothing. It means the plan sold on top of it was built before the evidence was.

What the biggest review actually found

The Sports Medicine meta-analysis is the one usually cited by both sides of this argument, so it is worth quoting rather than summarising.

Read the first conclusion again, because it is the argument in full. A meta-analysis over 78 studies did not conclude "train hard here and rest there". It concluded that general guidelines cannot be formed, and that the sensible response is a personalised approach. That is the opposite of a phase plan: a plan is a general guideline, applied to everybody, sold at scale.

Notice also what the credible interval does. Running from −0.16 to 0.04, it crosses zero, which means the possibility that early-follicular performance is slightly better sits inside the same range as the possibility that it is slightly worse. On low-quality evidence. That is not a foundation to reorganise a training year on.

Strength: the tidiest result in the field

Resistance training is where the claims tend to be boldest — build in one half, back off in the other — and it is where the evidence is least ambiguous.

When the review restricted itself to moderate-to-high-quality studies, 9 of 10 found no difference in strength performance between menstrual cycle phases. The authors' judgment is that it is premature to conclude that short-term fluctuations in reproductive hormones appreciably influence acute exercise performance or longer-term strength or hypertrophic adaptations.

Then they say the part that matters to anybody who has been handed a phase-based programme: recommendations to train by phase, as currently published, are premature and potentially misleading. Those are the words of researchers writing about their own field's advice, not the words of a website with an opinion about wellness. This page is simply repeating them.

The deeper problem: nobody knows which phase you are in

This is the cut that goes furthest, and it applies to this site as much as to anybody selling a programme.

A 2025 paper in Sports Medicine argues that assuming or estimating menstrual cycle phases has little scientific basis and lacks the rigour and methodological quality to produce valid and reliable data. Specifically: the calendar-based method — counting days between one period and the next — cannot be relied upon to determine a cycle or to classify its phases in research.

Sit with what that means. If a laboratory cannot trust a calendar to say which phase a participant was in, then an app cannot either, and neither can a printed chart, and neither can the arithmetic on this website. Athena's cycle phase calculator is exactly what that paper describes: two dates and a subtraction. It draws its boundaries as bands and says so on the page, which is honest, but honest arithmetic is still arithmetic. It is telling you what phase you are probably in, and "probably" is doing real work.

So a programme that assigns workouts to phases is stacking two problems. The effect it is training around is trivial and low-quality at best; and the input it uses to decide which day is which is a guess. Two soft numbers multiplied together do not produce a schedule. The evidence for where the boundaries actually sit, and how far the second half of a cycle moves, is in the luteal phase.

What the phases actually are

Describing the physiology is not the problem, and this article is not arguing that the cycle is uniform. A systems physiology review divides the cycle into two phases at ovulation — follicular and luteal — and subdivides those into early follicular, late follicular, ovulation, early luteal, midluteal and late luteal. The same review notes that cycle length can vary by up to 14 days between individuals and, less usually, within the same individual.

Six named subphases is a good map of a process. It is not a timetable, and the up-to-14-days line is the reason: the map's landmarks are in different places for different people, and sometimes in different places for the same person.

The line this site holds is the one between those two things. Describing what happens in a phase is physiology and belongs on a page like this. Prescribing a diet or a training block for each phase is a product. That is the distinction the whole of the period myths piece is built on, and it is the reason the word cycle-syncing appears on this page at all: naming the thing being refuted is the only way to answer the person who typed it into a search box.

Pain, and what this page cannot tell you about it

The one thing nobody disputes is that a lot of people are training through pain rather than around a phase chart.

More than half of women who menstruate have some pain for 1 to 2 days each month, and for some it is severe enough to keep them from normal activities for several days. ACOG explains the mechanism for primary dysmenorrhea: prostaglandins made in the lining of the uterus, at a high level on the first day of a period and falling as bleeding continues, which is why the pain tends to ease after the first few days. The NHS adds that period pain happens when the womb tightens, is often a normal part of the cycle, and usually lasts up to 3 days.

Here is where the advice actually inverts. Both publishers put exercise on the list of things that help. The NHS suggests gentle exercise — yoga, swimming, walking, cycling. ACOG goes further: exercising most days of the week can make you feel better, and aerobic workouts help produce chemicals that block pain. Neither of them says anything about which phase to do it in. The one place the cycle appears in a genuine exercise recommendation, it appears as keep going, and the days it names are the days a phase plan would have you back off.

Two things worth holding onto alongside that. Neither publisher is describing a treatment that removes the pain, and neither is a reason to train through something that is getting worse. The thresholds are the useful half: see a GP if your periods become more painful, heavier or irregular, or if the pain stops you doing your usual daily activities, and ask for an urgent GP appointment or help from NHS 111 if the pain is severe or worse than usual and painkillers have not helped. ACOG notes that secondary dysmenorrhea — caused by a disorder in the reproductive organs — tends to get worse over time and often lasts longer than ordinary cramps, which is why escalating pain is worth a doctor's time rather than a heavier warm-up. What else is on those lists, and who should not take an anti-inflammatory, is set out in period pain.

So what should you actually do

The meta-analysis already answered this, and its answer is the least marketable sentence in the field: a personalised approach. Not a phase template. Your own record.

That is duller than a plan and better than one. Log the session and log how it went, on the same page as your period start dates — a printable symptom tracker takes a row for energy or pain and a box a day, and a printable period tracker keeps the dates the pattern is measured against. After three or four months you will have something no general guideline can give you: evidence about you.

And if that record does show a repeating pattern — if the same two or three days are reliably worse every cycle — take it seriously as yours. Adjust what you do on the days your own log says are hard, in the way you would around a bad night's sleep or a heavy week at work. That is not a protocol and it should not be written as one, because it came from your data rather than from a chart printed for everybody. The same logic applies to sleep, where what women report and what a laboratory measures also come apart: your cycle and sleep sets out how far the evidence there actually goes.

The honest summary is almost boring, and it is nearly the opposite of what gets sold. Train. Train through most of it — that is the direction both health publishers point, and the only place the cycle enters their advice at all. Watch pain that is escalating rather than pain that is ordinary, and use the NHS's threshold for that rather than a feeling. Keep a record, because yours is the only dataset that is about you. And treat any confident schedule of what to do in each phase — cycle-syncing plans included — as what the strength reviewers called it: premature, and potentially misleading.

Two pages take this argument somewhere more specific. Sport and your period is for somebody with a fixture rather than a gym membership, including what to do when the fixture lands on a bad day. PE and periods is the same evidence at fourteen, where the real obstacle is usually white shorts and a changing room rather than a training plan.