Turtle data

Does training change across the cycle? Barely, our data shows

Turtle Method member data · Published 7 October 2026 · Next update April 2027 · Cite or embed this

Key findings

Comparing each Turtle member with herself, workout days stayed within 3 percentage points of her own average in every cycle phase (cycles from January 2025). Logged calories stayed within 10 kcal of her own average, with no premenstrual jump. Source: Turtle Method member data, turtlemethod.com/data/training-across-the-cycle/.

Under 3 pts

change in workout days in any cycle phase, against each woman's own average

Women kept training through every phase of their cycle. We compared each Turtle member with herself: the share of her app days with a workout in each phase, against her own average across the whole cycle. For more than 100 women, workout days stayed within 3 percentage points of their own average in every phase. The biggest change was a little more training during the period itself (+2.4 pts).

Workout days by cycle phase

Difference from each woman's own average share of days with a workout, in percentage points

Workout days by cycle phaseDiverging bar chart: every phase is within about 2 points of the average.Period (days 1 to 5)+2.4 ptsFollicular (day 6 to ovulation)−1.0 ptsLuteal (after ovulation)−0.6 ptsPremenstrual (last 6 days)−0.8 pts

Source: Turtle Method member data · CC BY 4.0

No premenstrual jump in logged food

Logged calories were just as steady. In every phase, women's logged calories stayed within 10 kcal of their own average, including the premenstrual days (+2 kcal). Logged protein stayed within 2 g.

Each phase against her own average
PhaseWorkout daysLogged caloriesLogged protein
Period (days 1 to 5)+2.4 pts+4 kcal−0.2 g
Follicular (day 6 to ovulation)−1.0 pts−10 kcal−1.7 g
Luteal (after ovulation)−0.6 pts+3 kcal+1.5 g
Premenstrual (last 6 days)−0.8 pts+2 kcal+0.5 g

Phases are a calendar estimate from logged periods (cycles of 21 to 35 days), not hormone tests: period = days 1 to 5, premenstrual = the 6 days before the next period, luteal = the 8 days before that, follicular = the days in between. Each measure covers more than 100 women with data in all four phases, so small differences are within the noise.

What the research says

  • Across 78 studies, performance might be trivially lower in the early follicular phase; the authors recommend a personalised approach rather than general rules.

    McNulty KL, Elliott-Sale KJ, Dolan E, et al. (2020). The effects of menstrual cycle phase on exercise performance in eumenorrheic women: a systematic review and meta-analysis. Sports Medicine. doi.org/10.1007/s40279-020-01319-3

  • An umbrella review concluded it is premature to say cycle phase appreciably affects strength or muscle gains from resistance training.

    Colenso-Semple LM, D'Souza AC, Elliott-Sale KJ, Phillips SM (2023). Current evidence shows no influence of women's menstrual cycle phase on acute strength performance or adaptations to resistance exercise training. Frontiers in Sports and Active Living. doi.org/10.3389/fspor.2023.1054542

Questions people ask

Do women train less before their period?

Barely, in Turtle's data: workout days in the premenstrual days were −0.8 pts against each woman's own average.

Does cycle syncing your workouts work?

Research finds only trivial differences in performance between phases, and recommends a personal approach. Train by how you feel on the day.

Method

Cycles of 21 to 35 days starting 1 January 2025 to 31 August 2026, from periods logged in the Turtle app; days up to 6 October 2026. Phases are a calendar estimate counted from the logged period start and back from the next one (period days 1 to 5, premenstrual the last 6 days, luteal the 8 days before that, follicular in between). For each woman and measure, the phase average minus her own average across all four phases; women count only with data in all four phases. Women only, staff accounts excluded; every group shown has 50 or more women and sizes are rounded. This shows a pattern, not a cause. Full methodology.

Cite this

Free to use with credit and a link to the source page. The figures and chart on this page are licensed under CC BY 4.0.

Journalists: for another cut of the data (aggregates only, same rules), contact scott@turtlemethod.com. More data on the Turtle data hub.

See whether your own training changes across your cycle with Turtle Method: Download for iPhone · Android