REF-1521 · Vivan et al., 2024 · International Journal of Sports Medicine · Paper · PubMed

Does easing off the first discipline pay?

In 38 triathletes, 20 km at 90% vs 80% FTP slowed men's 5-km run 35.1 s (women: no significant difference), yet bike + run was faster (57.7 s, 80.9 s). Triathlon only; no station tested.

Main caveat: Read from the abstract only. Level of athletes, test conditions, transition, 85% data and funding could not be checked.
moderate evidenceMeasured on recreational to highly trained athletes

Thirty-eight triathletes (21 men, 17 women) had body composition, VO2max and FTP assessed and then visited the laboratory three times to ride 20 km at 80%, 85% or 90% of FTP, in random order, and run 5 km as fast as possible.

Men ran faster after 80% than after 90% FTP (35.1 s; CI 2.2 to 68.1 s; p = 0.035), but their bike plus run was faster at 90% than at 80% (57.7 s; 26.1 to 89.3 s; p < 0.001). Women's run did not differ significantly by riding intensity, and their total was faster at 90% than at 80% (80.9 s; 29.7 to 132.1 s; p = 0.002).

The authors conclude that athletes should cycle at a minimum of 90% FTP. Our reading: within 80-90% of FTP over this short ride, riding harder paid in total time. The word "minimum" goes beyond the data, since nothing above 90% was tested and this adds nothing on where the run begins to collapse.

Protocol

Who: n = 38 triathletes, 21 men and 17 women; age, level and training are not in the abstract.
Conditions: 20 km at 80%, 85% or 90% of FTP, randomised order, then a maximal 5-km run. Bike type, indoor or outdoor, the run's surface and the transition are not in the abstract.
Baseline tests: body composition, VO2max, FTP.
Outcomes: 5-km run time; cycling plus running (C+R) time.
Funding and interests: funding is not in the abstract; the PubMed record prints "The authors declare that they have no conflict of interest."

Key Results

Comparison, 90% vs 80% FTPMen (n = 21)Women (n = 17)
5-km run35.1 s slower at 90% (CI 2.2 to 68.1 s; p = 0.035)no significant difference
Bike + run (C+R)57.7 s faster at 90% (CI 26.1 to 89.3 s; p < 0.001)80.9 s faster at 90% (CI 29.7 to 132.1 s; p = 0.002)

The 85% FTP results are not given in the abstract. The riding time saved is also not given; for men it would be about 92.8 s if the two differences can be added (our arithmetic).

Mechanisms

Measured: total and run times only, according to the abstract.
Authors' interpretation: none beyond the recommendation to ride at 90% FTP or more.
Inference: the arithmetic favours the longer leg. A given percentage of extra speed over the longer leg is worth more seconds than the same percentage over a 5-km run, so the result may shrink or reverse in a longer race, with more running, or when harder riding starts to wreck the run.

Practical Application

Does easing off the first discipline pay? In 38 triathletes, 20 km at 90% vs 80% FTP slowed men's 5-km run 35.1 s (women: no significant difference), yet bike + run was faster (57.7 s, 80.9 s). Triathlon only; no station tested.
Good for: a triathlete who rides a sprint-length bike leg far below threshold to save the run: over this range it did not pay on the combined time.
Not good for: first-kilometre pacing, which it does not test (REF-1508 does), or "a minimum of 90% FTP": nothing above it was tested, no swim preceded it, and no longer race was run.
HYROX (inference): not from this paper; the base's own practice: hold a reserve for the late stations (SYNTH-71). This study does not contradict that; it only warns that over-conserving has a cost, in a very different format.

Limitations

Read from the abstract only. Level of athletes, test conditions, transition, 85% data and funding could not be checked; the accepted manuscript at ZORA sits behind a bot check and was not tried further.
Sex-split analyses. Run difference shown in men but not in women; with 17 women, "no significant difference" is not the same as no effect, and no test of the sex difference is reported in the abstract.
Blinding is not mentioned in the abstract; the run is self-paced, so knowing the ride's intensity could shape it (inference).
Wide intervals. The men's run effect ranges from 2.2 to 68.1 s.
No swim, no race, one short distance. Conclusions about "triathlon performance" are extrapolated.
How this entry was checked and changed
Why this level

(from the abstract) — controlled crossover with n = 38 (rubric B+: ≥15), analysed by sex (21 men, 17 women), a performance outcome. The repository copy at ZORA answered with a bot check, so it was not read