SYNTH-37 · Guide

Coaching the menstrual cycle

The phase is not the lever; the symptoms, the heat and the athlete's own record are

Blimited evidenceMostly from recreational to highly trained athletesWritten before October 2026, being re-checked

Key Findings

1. Phase-based programming is not supported at population level. Neither endurance performance (REF-335) nor resistance-training response (REF-571) shows a reliable phase effect across women. "Train hard in the follicular phase, back off in the luteal phase" is a plan built on an effect the reviews find trivial (REF-335, low-quality evidence) or inconsistent (REF-571).

2. But the cycle is not invisible — it is thermal. The luteal phase raises core temperature before exercise in the heat with a large effect size (g 1.231) and keeps it elevated after exercise (g 0.455) (REF-709) — while sweat rate, skin temperature and exercise heart rate are unchanged. The cooling system works; the starting margin is smaller.

3. Whether that margin costs performance depends on the environment — and possibly on training status (an interpretation of two studies with different protocols, not a tested comparison). In recreationally active women, the luteal phase cost time to fatigue in hot-humid air (32°C/60% RH) but nothing at all in temperate conditions (REF-710). In well-trained women at matched WBGT, phase cost nothing while humidity cost 7% (REF-711).

4. What is linked to missed or changed training is symptom burden. In 6812 exercising women, 90.6% reported mood changes/anxiety and 86.2% fatigue, and a higher symptom index was associated with missing or changing training (OR 1.09 per index point), missing competitions (OR 1.07) and pain-medication use (REF-713). That survey did not measure phase; the phase null is REF-335 and REF-571.

**5.

Sex differences are a separate question from cycle phase; in thermoregulation they are smaller than commonly claimed.** At matched heat-loss requirements there was no sex difference in skin blood flow at any point; the sweating difference was not consistent at the lower loads (200 and 250 W·m⁻²), became significant at the highest (300 W·m⁻²), and came from lower output per sweat gland, not fewer active glands (REF-712, B, 8 men and 8 women).

Durability has one sex comparison in the base: after 90 min of unfuelled heavy cycling, power at the first threshold fell 2 ± 13% in women vs 16 ± 15% in men (sex × time P = 0.013), while peak power, VO2peak and neuromuscular function fell alike, individuals overlapped widely and the men did 25% more work per kg (REF-1132, B, 16 women and 16 men).

The women were tested in the early follicular phase or the pill- or ring-free week, or on a hormonal IUD or implant at any time (what the authors call a low-hormone state; no hormones were measured), so whether durability varies across the cycle is untested.

Practical Application

Default answer: do not restructure the week around cycle phase. Say why — the population evidence does not support it (REF-335, REF-571) — and say what you will do instead.
Ask about symptoms, then act on symptoms. Not from a study; the base's own practice: log a simple severity score alongside readiness and act on it rather than on a calendar rule — built on REF-713's association between the symptom index and missed or changed training. *Inference:* an athlete who regularly loses days to symptoms has a consistency problem worth more than any phase-based tweak could return.
Make the exception for heat. For a hard session or a race in hot, and especially humid, conditions during the luteal phase: buy back the thermal margin — earlier start, pre-cooling, in-session cooling, a more aggressive fluid plan — rather than cutting the session (not from a study; the base's own practice, built on the higher luteal core temperature with unchanged sweat rate in REF-709 and the hot-humid luteal deficit in REF-710).
Check the dew point before the calendar. Humidity produced a measurable performance decrement where phase did not (REF-711). When a hot session goes badly, that is the first variable to interrogate.
State which population your advice comes from. The luteal-phase deficit in the heat was found in recreationally active women and not in well-trained ones — do not quote one at the other.
Individual record beats group evidence here. Where group-level effects are absent but individual variation is large, the athlete's own tracked history across ≥3 cycles is the only instrument that can find her pattern (Not from a study; the base's own practice: the ≥3-cycle number. REF-335 recommends a personalised approach because group effects are trivial and variable; REF-336 recommends, in its full text, confirming regular cycles over several consecutive cycles; REF-337 supplies the individual-variation framing — none of the three abstracts carries "3").
Know the referral lines. Heavy bleeding → iron work-up (REF-573). Missing cycles → low energy availability until proven otherwise (REF-341). Severe or worsening symptoms → physician, not a coaching workaround.
Do not use the cycle as an explanation of last resort. A poor sweat response is not a phase effect — sweat rate, skin temperature and exercise heart rate did not differ by phase in the pooled data (REF-709; no performance outcome was pooled). The exception is hot-humid conditions, where REF-710 found a higher luteal heart rate and an earlier fade in recreationally active women.

Limitations

The evidence base under this framework is thin in absolute terms. REF-709 pools 9 studies and 83 participants; REF-710 tested 12 women per condition and analysed n=8 in each; REF-711 has n=10. The direction of the thermal finding is consistent, but the magnitudes are not settled.
The disagreement between [REF-710] and [REF-711] is attributed here to training status, which is an interpretation, not a tested comparison. Environmental differences (32°C/60% RH versus WBGT-matched 27°C) and exercise models differ too; no study has directly compared trained and untrained women under identical conditions.
Almost none of this applies cleanly to athletes on hormonal contraception, who were excluded from most of the underlying work (REF-338 is the relevant entry there).
[REF-713] is a self-selected self-reported survey and its symptom index is not psychometrically validated; it establishes an association (OR 1.09 per index point for missing or changing training), not a cause, and not how many days are lost.
Nothing here addresses perimenopause, adolescence, or athletes with gynaecological conditions — those are separate questions with separate evidence.
Symptom management is a clinical domain. This framework covers how to coach around symptoms, not how to treat them.
How this entry was checked and changed
Why this level

the phase null is A/B (REF-335, evidence quality graded low; REF-571, umbrella review), the heat findings rest on 9 pooled studies (n=83) and two trials of n=8-10, and the symptom, cooling and tracking rules are the base's own practice

Content correction 2026-09-27
"≥3 cycles (REF-336, REF-337)" -> the ≥3-cycle tracking standard is carried by REF-336 and REF-335; REF-337 carries only the individual-variation framing.All other cited numbers (REF-709 effect sizes 1.231/0.455, 9 studies/83 participants; REF-710 32°C/60% RH luteal deficit only in hot-humid; REF-711 7% humidity cost, no phase effect, n=10; REF-712 no skin-blood-flow sex difference; REF-713 6812 women, 90.6%/86.2%; REF-335 trivial early-follicular effect; REF-571 "premature to conclude") re-checked against the abstracts today and confirmed.
Sync after the full-base pass: "the ≥3-cycle standard is from REF-336 and REF-335" (bullet above) -> corrected REF-336 no longer carries a 3-cycle minimum (removed as not in the abstract; "several consecutive cycles" is full text) and REF-335 labels its three-cycle tracking as the base's own practice; relabelled house practice."REF-336/REF-337: how to track individually" -> coaching proxies are labelled the base's own practice in REF-336; hormone verification is REF-336 full text. REF-337 is now C+ (narrative methodology review).
Residual sync (REF-709/710/711/712/713 re-checked): REF-709 "core temperature raised at rest" -> pre-exercise (g 1.231) and post-exercise (g 0.455); heat-margin actions (pre-cooling, fluids, earlier start) labelled base practice as in REF-709. REF-713 "symptom burden, not phase, removes training days" -> the survey did not measure phase; symptom index associated with missing/changing training (OR 1.09 per point) and competitions (OR 1.07); "severity score beats any calendar rule" labelled base practice; the invented "2-4 days a month" removed (also removed from REF-713); "not by how much" -> per-point ORs given. REF-712 "sweating difference appears only at high heat loads" -> not consistent at 200/250 W·m⁻², significant at 300, from per-gland output (now B). REF-710 "n=12" -> 12 tested, n=8 analysed per condition.REF-711 claims match.
Resync 2026-09-28: re-read against REF-341 after its screening list was rebuilt from Table 4; "missing cycles -> low energy availability" still matches it.
Resync 2026-09-29: re-read against REF-341's relabelled caloric-restriction line; no claim here depends on it.
Verification 2026-10-01: all numbers check against the snapshot — g 1.231 / 0.455 (REF-709), 32°C/60% RH and n=12/8 (REF-710), 7% and n=10 (REF-711), 6812 / 90.6% / 86.2% / OR 1.09 / 1.07 (REF-713), 200/250/300 W·m⁻² and 8+8 (REF-712).
Verification 2026-10-01: the three prescriptive rules (symptom score, luteal heat cooling plan, >=3 cycles) are correctly labelled base practice; referral lines match REF-573 and REF-341 practice bullets.
Verification 2026-10-01: every cited entry has a DOI; no SOURCE flags.
Revision 2026-10-13: added REF-1132 (Pastorio 2026, full text) to Key Finding 5 and Cross-References; the Key Finding 5 title narrowed to thermoregulation ("smaller than commonly claimed" was shown for sweating and skin blood flow, not for durability). Level unchanged.