Short indoor interval training can improve durability about as much as long easy riding in less-trained adults, but it doesn't expose riders to the fatigue that appears after about two hours
Interpretation
One 10-week trial in 35 sedentary or recreationally active adults (16 M, 19 F) found 1.6 h/week of indoor intervals improved durability in a 3 h ride as much as 6.8 h/week of mostly outdoor low-intensity cycling (g 0.62 vs 0.49, p = 0.42 between groups) (Matomäki 2023). In 30 U23 professionals, more time below VT1 and a more polarised mix correlated with better power after 2,000 kJ (Spragg 2023). Fatigue at the first threshold and in critical-power markers appeared after about 2 h, not 40-80 min, in small, mostly male samples (Stevenson 2022; Clark 2019). No trial has compared indoor and outdoor long-ride training in trained triathletes.
Practical implication
Use indoor sessions for intervals and easy volume, keep an occasional ride of about two hours with fuelling practice through winter, and build back towards race-length rides in spring. This is inference from untrained adults, professionals and lab fatigue studies, not a tested triathlon plan.
Sources
6 sources, independently verified.
10 weeks of 6.8 h/week cycling below LT1 (mostly outdoors, long sessions 1.5-4 h) or 1.6 h/week indoor high-intensity intervals. Durability in a 3 h ride at about 48% VO2max (drift in HR, ventilation, energy expenditure, RPE and others) improved in both groups (g 0.49 and 0.62), with no between-group difference (p = 0.42). Interval group drift fell from 8.8% to 5.4% and onset moved from 108 to 137 min; low-intensity changes (7.7% to 6.3%, 106 to 131 min) were not significant. Only intervals raised VO2max. Previously untrained adults, low-intensity durability test.
Compared best 2, 5 and 12-minute power fresh (before 2,000 kJ) and fatigued (after 2,000 kJ) across a season. Fatigued 5 and 12-minute power was lower late season. More time below VT1 correlated with better fatigued 2-minute power (r = 0.43); a shift towards polarised intensity distribution correlated with better fatigued 12-minute power (r = 0.414). Correlational, young professionals.
Incremental step test before and after 2 h of cycling at 90% of estimated VT1. Power at VT1 fell from 217 to 196 W and at the lactate threshold from 212 to 190 W; heart rate at VT1 rose from 142 to 151 bpm. Attributed to reduced efficiency and altered energy expenditure. Small, almost all male.
3-minute all-out test fresh and after 40, 80 and 120 min of heavy-intensity cycling. End-test power 260 W fresh, 268 W at 40 min and 260 W at 80 min (no change), 236 W after 2 h with placebo (9% lower) and 254 W with 60 g/h carbohydrate. Work above end-test power fell from 17.9 kJ to 14.7 kJ at 80 min and 13.8 kJ at 2 h, not protected by carbohydrate.
Current-opinion paper defining 'durability' as the time of onset and magnitude of deterioration in physiological profiling characteristics (e.g. running economy, fractional utilisation) during prolonged exercise — the conceptual framework the rest of this batch's durability research sits inside. No abstract numbers in the indexed record; used only for the definition.
Randomised 20-minute time trials outdoors on a racing circuit and indoors on an electromagnetically braked ergometer, same bike and power meter. Mean power 280 W indoors vs 284 W outdoors, not different (P = .256), r = .94; within-person power variability far higher outdoors (SD 69 vs 33 W). All-male.