Cycling cadence (60, 80 or 100 rpm for 20 min) made no difference to a subsequent 3 km run in 9 well-trained triathletes
The run was 7-9% slower after any cycling than in isolation (626-638 s vs 583 s)
Participants
n=9 well-trained triathletes. Each completed a maximal cycling test, three cycle-run sessions (20 min cycling + 3000 m track run) in random order at 60, 80 and 100 rpm, and one isolated 3000 m run.
Cycling intensity approximated sprint-triathlon competition intensity (> 80% VO2max).
Protocol
Crossover with sessions in random order (from the abstract): each triathlete completed a maximal cycling test, three cycle-run sessions (20 minutes of cycling + a 3000 m track run) at 60, 80 and 100 rpm, and one isolated 3000 m run.
Cycling intensity approximated the athletes' sprint-triathlon competition intensity (> 80% VO2max).
Measures: 3000 m run time; stride rate and running velocity in the first 500 m; fraction of VO2max sustained during the run.
Key Findings
Prior cycling slowed the run, with no cadence effect: 3000 m times were 625.7 ± 40.1 s (60 rpm), 630.0 ± 44.8 s (80 rpm) and 637.7 ± 57.9 s (100 rpm) versus 583.0 ± 28.3 s isolated — the differences between cadences were not significant.
In the first 500 m of the run, stride rate and running velocity were higher after 80 or 100 rpm than after 60 rpm (p < 0.05).
60 rpm was associated with a higher fraction of VO2max sustained during the run than the other conditions (p < 0.05).
Authors' conclusion: running performance is altered after cycling, but no significant effect of cadence within the range triathletes usually use.
Practical Application
In 9 well-trained triathletes, 20 min of cycling at 60, 80 or 100 rpm produced statistically indistinguishable 3 km run times (626-638 s), all 7-9% slower than the 583 s isolated run — pick the cadence you race best at on the bike; do not expect a cadence change in the last minutes to protect the run.
The lower cadence (60 rpm) did produce a slower start (lower stride rate and velocity in the first 500 m) and a higher fraction of VO2max during the run — if anything, low cadence shifted cost into the run rather than sparing it.
Not from this paper; the base's own practice: rehearse your own bike-to-run cadence in brick sessions rather than assuming one setting fits everyone.
Limitations
Small sample (n=9), short 20-min bike bout and 3 km run; may not scale to Olympic/long-course loads.
Numerical trend toward faster runs after lower cadence was not significant and should not be read as an effect.
How this entry was checked and changed
Why this level
regraded 2026-09-27 from B+: crossover, n=9
Content correction 2026-09-27
Title and lead said "a lower final-lap cycling cadence tended to preserve subsequent 3 km run performance" and recommended dropping cadence -> abstract: "without any cadence effect"; no significant effect of cadence within the usual range. Recommendation removed.
Added the actual run times (625.7 / 630.0 / 637.7 vs 583.0 s), the first-500 m stride-rate/velocity finding (higher after 80/100 rpm) and the higher %VO2max after 60 rpm; the 7-9% figure is arithmetic on those times.
Level regraded B+ -> B (crossover, n=9).
2026-09-28: added Protocol from the abstract (Europe PMC PMID 12663359).