Punch Force and Explosive Power Peaked at Ovulation in 24 Elite Female Boxers

A Chinese research team tested 24 elite female boxers at three points in the cycle and found that menstrual cycle phase tracked with punch force and explosive power. Hormone levels, however, did not line up with the performance swings.

Menstrual cycle phase was associated with measurable differences in straight punch performance and lower-body power among elite female boxers, according to an Original Research paper accepted on October 29, 2026 in Frontiers in Physiology. Every punch and power variable the team measured peaked in the ovulatory phase.

The study comes from Wuhan Sport University, with co-authors at Guangxi Medical University and the Wuhan Institute of Technology. Tian Tian, Xingran Liu, Feiyang Yu, Lu Zhou and corresponding author Shanjun Bao recruited 24 national first-class female boxers and tested each of them repeatedly rather than splitting them into groups.

That prospective, repeated-measures design means every athlete served as her own comparison across three phases: early follicular, ovulatory and mid-luteal. Phase assignment was not left to self-report. The team verified it with basal body temperature, urinary luteinizing hormone testing and salivary hormone assays.

Across the battery, the reported partial eta squared values ranged from 0.175 to 0.451, and between 70.8% and 91.7% of the athletes improved during the ovulatory phase depending on the measure. There were no significant differences between the early follicular and mid-luteal phases.

How the Menstrual Cycle Phase Study Was Run

The outcome list is unusually specific for this literature. Rather than relying on a single jump test, the team measured lead and rear straight punch peak force and peak velocity, countermovement jump relative peak power, rotational medicine ball throw distance and bench press throw relative peak power.

Hormonal work used liquid chromatography tandem mass spectrometry to quantify salivary estradiol, progesterone and free testosterone. The authors report that hormones fluctuated as expected across the cycle, which they treat as evidence that menstrual cycle phase classification in the sample was valid.

Statistical handling used repeated-measures analysis of variance with Bonferroni-adjusted post hoc comparisons, partial eta squared as the effect size, individual response proportions and correlations between hormone change and performance change.

That verification step is worth dwelling on. Much of the earlier literature in this area assigned menstrual cycle phase from calendar counting alone, which misclassifies a sizeable share of cycles because ovulation timing varies. Combining temperature, urinary luteinizing hormone and salivary assays is a stricter standard than most comparable studies have met.

Every Power Measure Peaked in the Ovulatory Menstrual Cycle Phase

The menstrual cycle phase pattern was consistent rather than scattered. Punch force, punch velocity, jump power, medicine ball throw distance and bench press throw power all registered their highest values in the ovulatory phase. Partial eta squared values between 0.175 and 0.451 describe moderate to large effects by conventional benchmarks.

Just as notable is what did not differ. The early follicular and mid-luteal phases did not separate from one another statistically. In this sample the useful contrast was ovulatory against the rest, not a smooth gradient running through the cycle.

Partial eta squared describes how much of the variance in a measure the menstrual cycle phase factor accounts for. A value of 0.451 at the top of the reported range is substantial for a single factor in applied sport science, though with 24 athletes and no confidence intervals the plausible range around each estimate stays wide.

Hormone Levels Did Not Track Performance

Here the study gets more interesting than a simple ovulation-is-better headline. Correlations between individual hormone changes and individual performance changes were not significant, with Pearson r values spanning minus 0.180 to 0.395 and all p values above 0.05.

So the group-level timing effect held while the proposed mechanism did not show up at the individual level. Knowing an athlete’s estradiol or progesterone reading on a given day did not predict whether her punch would land harder. That decoupling is the authors’ main reason for caution.

It also means menstrual cycle phase in this dataset behaves as a timing marker rather than as a measured hormonal cause. Something that co-varies with phase may be driving the differences, and salivary hormone concentrations are not obviously it.

Why Menstrual Cycle Phase Response Varied So Much

Between 70.8% and 91.7% of athletes improved in the ovulatory phase, which is the same as saying that for some outcomes close to three in ten did not. Individual variability on this scale is a recurring feature of menstrual cycle phase research, and it is why the authors push back on blanket programming.

A squad-wide rule built on a group average would therefore misdirect a meaningful minority of athletes. The practical implication the authors draw is individual monitoring: establish whether a given athlete shows a menstrual cycle phase pattern before assuming she does.

What the Menstrual Cycle Phase Findings Do Not Establish

This was an observational repeated-measures study, not a training intervention. Nobody trained differently by phase, and no competitive outcome was measured. The study does not show that scheduling hard sessions around ovulation improves adaptation, performance in a bout or anything else over a season.

It also does not speak to athletes on hormonal contraception, to non-elite boxers, or to sports with different power demands. And because effect sizes arrive without confidence intervals in the accepted version, the precision of each menstrual cycle phase estimate cannot be judged from what is published so far.

Nor does a menstrual cycle phase difference in a laboratory test necessarily survive contact with a real bout, where pacing, tactics, opponent and fatigue swamp small physical margins. Translating a force-plate advantage into a competitive one is a separate research question that this design was not built to answer.

How Menstrual Cycle Phase Monitoring Could Be Approached

None of the following is individualized medical or coaching advice, and anyone with cycle-related symptoms or performance concerns should speak with a qualified clinician. As general, low-risk practice, the study points in a clear direction for athletes who want to look at their own data.

• Track simple performance markers alongside cycle dates for several months before drawing any conclusion about menstrual cycle phase effects.
• Use measures you can repeat reliably, such as a jump test or a timed throw, rather than subjective feel alone.
• Expect either a personal menstrual cycle phase pattern or none at all; both were present in this sample.
• Treat a single strong or weak session as noise, not as evidence.

The caution about small samples is familiar from earlier FLM coverage. A separate trial reported that caffeine made 30-meter sprints about 1.5% faster in a small female athlete trial, and a scoping review found that countermovement jump fatigue testing varied widely across 117 studies, which complicates comparison between any two power datasets.

On the combat sports side, a study of reaction timing errors in 312 combat sports athletes shows how much within-athlete variation these disciplines carry, and work on resistance priming and jump height at 30 hours is a reminder that timing effects in power output are not unique to menstrual cycle phase research.