Polarized Training Has Greater Impact on Key Endurance Variables than Threshold, High Intensity, or High Volume Training
Participants
n=48 well-trained endurance athletes (runners, cyclists, triathletes, cross-country skiers); VO2peak 62.6 ± 7.1 mL/kg/min
Protocol
9 weeks, random assignment to 4 groups: high-volume training (HVT), threshold training (THR), high-intensity interval training (HIIT), polarized (POL)
Training intensity heart-rate controlled
Pre/post: incremental test, work economy, VO2peak test
Zone split per group (full text, checked 2026-09-27): HVT ~83/16/1%, THR ~46/54/0%, HIIT ~43/0/57%, POL ~68/6/26% (zone 1/2/3)
Key Results
POL → greatest improvements: VO2peak +6.8 mL/kg/min (+11.7%, P<0.001), ramp time to exhaustion +17.4% (P<0.001), peak velocity/power +5.1% (P<0.01)
Velocity/power at 4 mmol/L: +8.1% after POL (P<0.01) and +5.6% after HIIT (P<0.05)
HIIT: VO2peak +4.8 ± 5.6% (P<0.05) and time to exhaustion +8.8% (corrected 2026-10-12, full text); body mass −3.7% (P<0.001); no body-mass change in other groups
THR and HVT: no further improvement in any measured endurance-performance variable (P>0.05), apart from a slight improvement in work economy in THR
Work economy: no between-group differences in pre-post change
Mechanism
Inference, not tested: the high-volume low-intensity part of POL supplies peripheral (mitochondrial/capillary) stimulus while the high-intensity part supplies the VO2max stimulus; chronic threshold work may add fatigue without enough of either. The paper did not measure these mechanisms.
Practical Application
In n=48 well-trained endurance athletes (VO2peak ~63) over 9 weeks, polarized training raised VO2peak by 6.8 mL/kg/min (+11.7%), ramp time to exhaustion by 17.4% and peak power/velocity by 5.1%, while threshold-dominant and high-volume training brought no further gains in performance variables — for trained athletes, a mix of mostly easy work plus a real block of high-intensity work beats a steady diet of threshold.
POL in this study was roughly 68/6/26% by zone (full text, checked 2026-09-27), i.e. more high-intensity time than the popular "80/20" label suggests.
Inference, not tested (the Norwegian method was not studied here): the Norwegian method differs from the "threshold" group here — LGTIT uses sub-threshold intervals, not continuous threshold work; these findings argue against excessive continuous zone-3 work rather than against the Norwegian approach.
Key nuance: Bakken's "golden zone" (sub-threshold) may avoid the fatigue trap of the THR group (inference, not tested).
Limitations
9 weeks — may not reflect long-term adaptations
Mixed-sport population (not sport-specific)
"Threshold" group trained AT threshold, not BELOW it as in Norwegian method — important distinction
How this entry was checked and changed
Why this level
regraded 2026-09-27 from A: parallel RCT, n=48 across four groups, not pre-registered
Content correction 2026-09-27
"VO2max +6.8%" -> VO2peak +6.8 mL/kg/min, i.e. +11.7%.
Invented THR (+4.5%, TTE +8.8%, "second best") and the HIT TTE +2.7% and the "composite score" deleted (the HIT VO2peak +4.8% was real: +4.8 ± 5.6%, P < 0.05; HIIT TTE was +8.8% (corrected 2026-10-12, full text)) -> THR and HVT gave no further performance gains; HIIT raised v/P at 4 mmol/L +5.6% and cut body mass 3.7%.
Lead said "80/20" -> POL split was ~68/6/26% (full text); Sport line adds xc-skiing.
Level A -> A- per the level rubric (parallel RCT with n=48 in four arms; A needs a large pre-registered trial or a systematic review).
Lines labelled "full text; not checked against the abstract" were checked against the open-access full text (PMC3912323): every number in them appears there; relabelled "full text, checked 2026-09-27".
Resync 2026-09-28: Norwegian-method comparison stated as a finding -> labelled inference.
Full text (checked 2026-10-12)
Participants were male and female athletes aged 31 ± 6 years (8 cross-country skiers, 21 runners, 4 triathletes, 15 cyclists) training 10–20 h·wk−1; none had regularly done HIIT in the previous 6 months, and all had followed HVT with a maximum of two THR sessions per week (full text).
POL hard weeks had six sessions: two 60 min HIIT sessions, two 150–240 min LOW sessions including six to eight maximal 5 s sprints separated by at least 20 min, and two 90 min LOW sessions; the recovery week had one 60 min HIIT, one 120–180 min LOW and one 90 min LOW session (full text).
Every HIIT session was a 20 min warm-up at 75% of HRpeak, 4 × 4 min at 90–95% of HRpeak with 3 min active recovery, and a 15 min cool-down (full text).
THR hard weeks had two 60 min LT interval sessions (5 × 6 min with 2 min recovery, then 6 × 7 min, then 6 × 8 min by block), one 90 min LT session (3 × 15 min with 3 min recovery, then 3 × 20 min), one 75 min fartlek and two 90 min LOW sessions; zones were LOW <2 mmol·L−1, LT 3–5 mmol·L−1 and HIGH >90% HRpeak (full text).
41 of 48 completed; seven withdrew or were excluded (2 in HIIT, 1 in HVT, 4 in THR), and total training time was 104 ± 20 h for POL, 66 ± 1 for HIIT, 84 ± 7 for THR and 102 ± 11 for HVT (full text, Table 1).
VO2peak in POL rose from 60.6 ± 8.3 to 67.4 ± 7.7 ml·min−1·kg−1, HIIT rose 4.8 ± 5.6% (P < 0.05), and individual changes in THR ranged from −20 to +4% (full text).
Time to exhaustion rose 17.4% in POL, 8.8% in HIIT, 6.2% in THR and 8.0% in HVT, with no statistical differences between groups (full text, Table 3).
Correction: the entry's content-correction note lists "HIT (+4.8%, ...)" among invented numbers, but the paper reports a 4.8 ± 5.6% VO2peak increase after HIIT (P < 0.05); the HIIT time-to-exhaustion gain was +8.8% (full text).