What is claimed
The story goes roughly like this: in the first half of the cycle oestrogen rises, so that's the time for heavy weights and hard intervals. In the second half progesterone dominates, so recovery belongs there. This turns into training plans that tell you each week what your body supposedly needs.
It sounds plausible and pleasantly concrete. Since 2026 there has been a direct test of it.
The direct test: a 2026 randomised trial
A research group at McMaster University had 24 women do strength training over three cycles, around twelve weeks. Each leg was randomly assigned its own variant: no training at all, even training across the whole cycle, high volume in the follicular phase and low volume in the luteal phase, or exactly the reverse.
The result: all trained variants gained clearly in muscle mass and strength, and all by the same amount. Whether the volume fell in the first or second half of the cycle made no difference. What was linked to the gains was total training volume. The authors note: if you want to plan by phase, you can, but it isn't necessary for building muscle.
What the reviews show
A 2025 systematic review only considered studies in which hormones were actually measured in the blood – 19 studies with 279 women in total. Maximal and explosive strength stayed largely the same across the cycle. Some studies found slightly lower endurance or peak performance in the first days of the period and slightly better coordination around ovulation. However, the studies' risk of bias was moderate to high.
A 2026 meta-analysis of elite athletes found no significant difference in maximal oxygen uptake between phases, only a small, non-significant downward trend during the period. An assessment of reviews spanning 40 years reaches the same picture: most find no effect or only a small one.
The long most-cited 2020 meta-analysis had already hinted at this – very small effects with mostly low study quality. The newer work with measured hormone levels confirms it.
Three reasons why older findings were so inconsistent:
- The samples are small. Many studies work with a double-digit number of participants. With small effects, that isn't enough to separate chance from signal.
- The phase is often estimated rather than measured. If ovulation isn't confirmed by hormone testing but derived from the calendar, some measurements are assigned to the wrong phase. The result stands or falls with that.
- Definitions differ. “Follicular phase” means the days right after the period in one paper and the week before ovulation in another. Such studies can only be pooled to a limited extent.
What counts instead
A second 2026 study followed 28 trained women over two training blocks and measured bar speed on every repetition. The cycle phase changed performance only minimally. What was clearly linked to performance, by contrast, was how motivated the women were on that day.
The sports nutrition society ISSN accordingly recommends that female athletes observe their own patterns across the cycle instead of following a general rule.
What we built from this
That's why, since version 1.6, Rhythmus no longer gives training instructions per cycle phase. Your weekly plan stays as you set it. For each phase, the app shows what research knows, with a rating and a source.
You log daily how you feel; if the check-in is low, the app suggests a gentler session – based on your own entry, not a model. And after a few cycles, “Your pattern” shows whether something recurs at the same point in your own cycle. The analysis accounts for weekdays and only reports what is confirmed twice in a row.
That's the less spectacular option. Its advantage is that it also works if your cycle is irregular, if you use hormonal contraception, or if you simply are one of the women in whom the described patterns don't appear.