REF-502 · Roete, van Erp, Elferink-Gemser, Kempe & Lamberts, 2025 · International Journal of Sports Physiology and Performance, · Paper
Internal and External Load Characteristics in Professional Female Cyclists During Highly Successful and Less Successful Seasons
Participants
n=14 female World Tour cyclists (age 24 ± 4 y), 43 seasons in total (3 ± 1 per rider) over a 7-year period — 9,036 training sessions and 1,974 races. Seasons classified by ProCyclingStats points per race: "highly successful" (≥5 points per race) vs "less successful" (<5).
Protocol
Exploratory longitudinal observational design comparing each rider's own seasons: forty-three seasons of 14 World Tour cyclists over a 7-year period (from the abstract).
Seasons classified by ProCyclingStats points per race as "highly successful" (≥5 points per race) or "less successful" (<5 points per race).
External and internal training and racing load characteristics and intensity distributions (power zones) compared between season types with a multilevel random intercept model followed by pairwise comparison.
Key Findings
Most external load characteristics — distance, duration, kilojoules — were HIGHER in highly successful seasons (P < .01), while overall mean power output was similar.
In highly successful seasons riders trained more in power zones 1 and 2 (P < .05); in less successful seasons they spent relatively more race time in zone 5 (P < .01).
No differences in internal load characteristics between season types — the signal was in external load only.
Authors' conclusion: in highly successful seasons the riders trained more (volume) at a lower relative intensity in both training and racing; the findings "highlight the importance of high-volume low-intensity training", which may contribute to better durability in races.
Practical Application
Roete et al. 2025 (14 World Tour women, 43 seasons): the seasons that scored were the ones with more total riding — longer, further, more kilojoules — and more time in power zones 1-2, at a similar mean power; internal load did not differ. At elite level the better year was the bigger, easier-on-average year, not a harder one.
The within-athlete design compares each rider's own seasons, which removes trait-level differences (talent, team, genetics) from the comparison — but it is observational and exploratory; it cannot say the volume caused the results.
Inference, not tested: for long-format goals the pattern argues for adding low-intensity hours before adding intensity; the study did not test a prescription and gives no hours figure.
Mechanism (Inference, not tested — consistent with [REF-465] Mølmen 2025)
Not from this paper: REF-465 reports that endurance training and HIT improved more slowly but steadily over a greater number of training weeks (it reports no HIT plateau). That the low-intensity layer keeps paying across a long competitive season is consistent with that, and reading this study as within-athlete confirmation of it is the base's inference; the abstract itself points to durability.
Limitations
Observational and exploratory — cannot establish causality; 14 riders
Female World Tour cohort — generalisation to male, masters, amateur populations should be cautious
"Success" = ProCyclingStats points per race (≥5 vs <5) — a results metric, not a physiological one
Does not decompose WHICH low-intensity sessions or formats were associated with success — volume and zone-distribution analysis
Cannot separate the volume effect from other factors (health, injury, team role, sleep, nutrition) that may co-vary across successful seasons
How this entry was checked and changed
Why this level
regraded 2026-09-27 from A−: observational cohort of 14 riders / 43 seasons, exploratory by the authors' own description
Content correction 2026-09-27
Source verified through the base's own DOI map (10.1123/ijspp.2024-0365; no DOI line added to the entry). Level A- -> B: exploratory observational cohort, 14 riders / 43 seasons. Sport "cycling, running, swimming" -> cycling.
"Only Z1 volume differentiated seasons; total volume, HIT, MIT, racing volume did not" -> the abstract says most external load (distance, duration, kJ) was higher in successful seasons and riders trained more in zones 1 AND 2, with more race time in zone 5 in the less successful seasons.
"Internal load confirmed the pattern" -> the abstract reports NO differences in internal load. Added n=14, 43 seasons, 9,036 sessions, 1,974 races, the ≥5 points/race definition; mechanism relabelled as inference from REF-465; REF-197 byline 2025 -> 2024.
Mechanism: "HIT gains flatten around weeks 15-20 while low-intensity gains continue" was attributed to REF-465, which reports no HIT plateau; restated to what REF-465 says (slower but steady gains for ET and HIT).
Cross-references to duplicate stubs repointed to their canonical entries (REF-085->REF-303); the stubs were emptied on 2026-09-27.
2026-09-28: added Protocol from the abstract (Europe PMC, DOI 10.1123/ijspp.2024-0365 as recorded in this entry's 2026-09-27 correction note, PMID 40681164).
2026-10-01: cross-reference repointed from a merged framework to the one that replaced it (SYNTH-01→31, 05→34, 09→70, 12/15→32, 19→26).
2026-10-01: citation of the superseded pointer SYNTH-19 repointed to SYNTH-26, which carries the cited content.