Is the AT algorithm training specifying for female-bodied athletes?

Hey TR team, I’ve been watching the AT in action for my wife, who is a PE teacher and leads the education and training of female-bodied athletes for our school’s athletes, and it seems to me that the AT is missing a massive input for females - menstruation.

The rapid growth in recent literature shows menstruation has a range of effects on training, from nutrition requirements, and responses to different stress inputs, to susceptibility to RED-S, pre-during-post workout fueling, and that more research is needed:

Vogel, Kurt, et al. “Female Athletes and the Menstrual Cycle in Team Sports: Current State of Play and Considerations for Future Research.” Sports 12.1 (2024): 4. ProQuest. Web. 5 Feb. 2024.

Niering, Marc, et al. “The Influence of Menstrual Cycle Phases on Maximal Strength Performance in Healthy Female Adults: A Systematic Review with Meta-Analysis.” Sports 12.1 (2024): 31. ProQuest. Web. 5 Feb. 2024.

McGawley, K., et al. “Improving Menstrual Health Literacy in Sport.” Journal of Science and Medicine in Sport 26.7 (2023): 351-7. ProQuest. Web. 5 Feb. 2024.

I’m really interested to see how this one turns out: Ekenros, Linda, et al. “Impact of Menstrual Cycle-Based Periodized Training on Aerobic Performance, a Clinical Trial Study protocol—the IMPACT Study.” Trials 25.1 (2024): 93. ProQuest. Web. 5 Feb. 2024. Especially from this one:

The impact of the menstrual cycle on training and physical performance is still not elucidated. Although previous studies on periodized training indicate an eventual advantage for follicular-based training [3], the results are not conclusive. Furthermore, the role of menstrual cycle-related symptoms for training and performance has not been fully explored. A recent study suggested that motivation to exercise is of superior significance rather than variation in hormone levels during the menstrual cycle to improve physical performance [4]. As several menstrual cycle-related symptoms such as dysmenorrhea and PMS are common among athletes , these symptoms might affect motivation and the effectiveness of training sessions.

And the above is just from the first 20 results in ProQuest for ‘menstruation endurance athlete’, I’ve barely skimmed the surface.

There’s a genuine opportunity for TR to lead the pack (pun intended) by integrating established research into the AT algorithm for female-bodied athletes - the dataset you’d generate would produce some really interesting insights, and the usefulness (from adapting individual workouts to building plan calendars appropriately) would be unique amongst training platforms. Many female-bodied athletes are not comfortable discussing menstruation with their (usually male) coaches, so incorporating this data into the AT algorithm would be an awesome stopgap while social norms catch up.

I’d love to see this topic brought up on the podcast, too - and revisited as new literature emerges.


Great post! I agree that this would be great, and I’m pretty sure I’ve heard it mentioned on the podcast as something that is planned.

Nate said on the podcast (maybe last Summer?) that they took advantage of the new Apple Health integrations to begin gathering this data. I think if you log that info into Apple Health then the TR platform does prompt you to opt-in to sharing it with them.

IIRC they aren’t doing anything with the data just yet. I’d guess they’re probably trialling some ML models in early development stages and once they feel like they have a big enough dataset to make meaningful and predictable adjustments then it will get folded into AT somehow.

That says to me that it could be straightforward and quick if the correlations are really distinct and consistent or that it could just get continually postponed like WLV2 if things are more nuanced. Either way it sounds like it’s in the product road map and first steps are in place but it’s still a fair way away from making it to us.

(Caveat: whilst I’ve listened with great curiosity I’m cis-male so have no direct experience of any of this.)

1 Like