REF-1240 · Mateo-March et al., 2025 · Journal of Science and Medicine in Sport · Paper · PubMed

Do women fade more than men late in a race?

In 42 female and 42 male pros' files, after equal work per kg, women's best 1-20 min power fell 2-8 points more; a time-matched runner study found the reverse. No sex rule.

Blimited evidenceMeasured on elite to world-class athletes

Mateo-March et al. asked whether durability differs between female and male professional cyclists. From power files of training and competition over 1-5 seasons, they took each rider's highest power for 10 s, 1 min, 5 min and 20 min (the record power profile) when fresh (0 kJ/kg) and after 10, 20 and 30 kJ/kg of accumulated work, in 42 women and 42 men.

Record power fell significantly in both sexes only after more than 10 kJ/kg (P < 0.001), with no significant fall below that. The relative decline in 10-s power did not differ between the sexes, but after 20 kJ/kg the women's decline was larger for 1-min, 5-min and 20-min efforts (by 4, 4 and 2, written as %, P < 0.05), and after 30 kJ/kg larger still (8, 6 and 7, P < 0.001).

The authors conclude that professional women show a greater relative decay after a given amount of work, which "might reflect a lower durability".

Our reading: a real field signal at matched work per kilogram, pointing the opposite way from the time-matched running study (REF-1241) and from the threshold result of the cycling one (REF-1132, where peak power fell alike in both sexes); the matching basis alone could account for much of the disagreement, and the abstract gives none of the descriptive data needed to judge race patterns or sampling.

Protocol

Design (from the abstract): observational, field-based; power output from training and competition.
Who: n = 84 professional cyclists, 42 women and 42 men, each with 1-5 seasons of data.
Outcome: the record power profile, the highest power for 10 s, 1 min, 5 min and 20 min, under non-fatigued conditions (0 kJ/kg) and after 10, 20 and 30 kJ/kg of accumulated work.
Comparison: the relative decay (% decline from fresh) between sexes at each work level.
Not given in the abstract: team categories and racing calendars, age and fresh power values, the number of files per rider and per work level, the share of races vs training, how accumulated work was computed, the intensity at which the work was done, the statistical model, and the absolute declines in each sex. Funding listed in Crossref and Europe PMC: Instituto de Salud Carlos III (Europe PMC grant CD21/00138); conflicts not given in the abstract.

Key Results

Accumulated workBoth sexes vs freshWomen's extra relative decline vs men (abstract writes %)
10 kJ/kgno significant fall (P > 0.05 for all)not reported as different
20 kJ/kgsignificant fall (P < 0.001)10 s: similar · 1 min: 4 · 5 min: 4 · 20 min: 2 (P < 0.05)
30 kJ/kgsignificant fall (P < 0.001)10 s: similar · 1 min: 8 · 5 min: 6 · 20 min: 7 (P < 0.001)
Reading the units: the abstract gives the sex differences as "4 %, 4 % and 2 %" and "8 %, 6 % and 7 %" in relative decay; these are most plausibly percentage points of decline, and the per-sex declines are not given.
Onset: neither sex showed a significant fall at 10 kJ/kg; the falls and the sex gap appear from 20 kJ/kg, in line with the ~20 kJ/kg onset in professionals' files in REF-479.

Mechanisms

Measured: at equal kJ/kg, professional women's best efforts of 1-20 min had fallen further from their fresh bests than men's; 10-s efforts fell alike.
Authors' interpretation: a greater relative decay after a given accumulated work, which might reflect lower durability in female professionals.
Inference (matching basis): at equal kJ/kg a rider with lower power per kilogram has ridden longer, so a work-matched comparison gives women more time; the two time-matched laboratory studies gave the men more work per kilogram (25% more work per kilogram in REF-1132; in REF-1241 the women spent 16% less energy per kilogram). Each design hands the bigger dose to the other sex.
Inference (what kJ hides): the review co-written by the first author found that work above CP costs far more later power than the same kJ below it (REF-1239); if the share of work above CP at a given kJ/kg differs between women's and men's racing, kJ/kg matching confounds sex with intensity.The abstract does not report intensity; the REF-1241 authors cite a later study by the same first author (Int J Sports Med 2026, not in the base, not read) that matched riders by work above CP and still found men more durable, so this confound may not explain the gap.
Inference (record sampling): a record after 30 kJ/kg is the best effort among the files that reached 30 kJ/kg; if fewer of one group's files, or fewer of its decisive race moments, fall beyond that point, its record there is a best of fewer chances and lower partly for that reason. The abstract gives no file counts, so this cannot be checked.

Practical Application

Do women fade more than men late in a race? In 42 female and 42 male pros' files, after equal work per kg, women's best 1-20 min power fell 2-8 points more; a time-matched runner study found the reverse. No sex rule.
Both directions are now in the base: time-matched laboratory studies favoured women (REF-1132 cycling, unfuelled: thresholds slid less while peak power and VO2peak fell alike; REF-1241 running, fuelled: uphill time-trial speed fell less), and this work-matched field study found female professionals faded more.No sex rule follows; this supports SYNTH-32's line not to assume a woman will fade less because of her sex, and equally argues against assuming she will fade more.
Read her own data: compare a woman's fatigued powers with her own fresh ones at the work levels she actually races, using her own repeated files (REF-479); do not import men's percentages or a sex correction (inference). The group gap at 20 kJ/kg (2-4 points) is of the same order as the measurement error of one professional's fatigued records (SEM under 5% in REF-479), so one rider's file cannot show it (inference).
Plan late efforts from her fatigued records, not her fresh ones: after 20-30 kJ/kg both sexes were below their fresh bests, and the women by more; set late-race targets from her own records at the work her races reach (inference).
Why the studies may differ: matched work vs matched time; field records vs set laboratory tests; professional road racers vs highly trained amateurs; cyclists vs runners (one of the two laboratory studies); uncontrolled race fuelling vs water only (REF-1132) or 90 g/h (REF-1241); race-like intensity vs steady moderate or heavy riding.

Limitations

Abstract only. The full text (Elsevier) was not open access and OpenAlex lists no repository copy; descriptive data, file counts, team levels and per-sex declines are not given in the abstract.
Observational field records. The efforts were whatever races and training produced; tactics, race length, terrain, heat and fuelling were not controlled, and a "record" depends on how many chances a rider had at each work level.
Matched by kJ/kg, not by time or intensity, which by construction gives the lower-watt group more riding time per bin (inference); a review co-written by the first author argues that kJ alone ignores intensity (REF-1239).
Professionals only, in a sport where women's and men's racing differ in distance and format; transfer to amateurs or other sports is untested.
Same research group as REF-192 and REF-479, which use the same record-power method, and as the review REF-1239 (shared authors).
Reporting: the size of the sex gap is given only as a difference, in units written as % that are probably percentage points.
How this entry was checked and changed
Why this level

from the abstract; observational cohort of n = 84 (42 per sex; under the rubric's 200 for B+), exposure as it happened in races and training rather than a set protocol, outcome the best power produced after each work level rather than a test; per-sex declines, file counts and team levels are not given in the abstract