Polarised or pyramidal: zones, counting and updated evidence
On this page
Three physiological zones, not your device zones
| Zone | Boundary in the three-zone model | Caution |
|---|---|---|
| Z1 | Below first ventilatory or lactate threshold | Not automatically Z1 on a five-zone watch |
| Z2 | Between first and second thresholds | Not the “easy zone2” in every commercial scale |
| Z3 | Above second threshold | Does not specify interval duration or recovery |
Zone numbers depend on the model. Comparing percentages requires matching boundaries and measurement methods. Talking and perceived effort help in practice but do not replace threshold measurement when precision is needed. Marathon pace is not automatically assigned to the same zone for every runner.3
Pyramidal and polarised
In a pyramidal distribution, Z1 has the greatest share, followed by Z2 then Z3. In a polarised distribution, Z1 still dominates but Z3 exceeds Z2: the middle share is reduced. These labels describe an observed period; they do not require identical percentages every week.3
| Arithmetic example only | Z1 | Z2 | Z3 |
|---|---|---|---|
| 300 pyramidal minutes | 240 min (80%) | 45 min (15%) | 15 min (5%) |
| 300 polarised minutes | 240 min (80%) | 15 min (5%) | 45 min (15%) |
Both examples have 80% easy work but different stimuli: 80/20 alone does not distinguish pyramidal from polarised. These numbers explain distributions; they are not individual targets or loads to adopt.
Minutes and sessions are not interchangeable
In 2010, Seiler described national and international athletes completing 10–13 weekly sessions: about 80% of sessions were easy and 20% dominated by hard work. That does not prescribe 20% of minutes above threshold for everyone.1
A five-session example
Four easy sessions and one interval session are 80/20 by session count. If all last 60 minutes, the total is 300. If the hard session includes 15 minutes above the second threshold and 45 easy minutes, total time is 285 minutes Z1 and 15 Z3: 95/0/5, not 80/0/20. Warm-up, recoveries and cool-down change the count. Heart-rate counting is also affected by its delayed rise during short repeats.
State whether you count sessions by main purpose, time in heart-rate zones, or time at a given speed/power. Do not merge results calculated with different criteria.
What the research shows
| Source | Sample and design | Result to interpret |
|---|---|---|
| Stöggl and Sperlich 2014 | 48 trained multi-sport athletes, 9 weeks; four groups | POL: greatest VO₂peak increase (+11.7%) and other variables; no between-group economy difference |
| Oliveira et al. 2024 | Meta-analysis:17 studies, 437 participants | Small POL advantage for VO₂peak; no superiority found for time trial, time to exhaustion or speed/power at second threshold |
| Rosenblat et al. 2025 | Individual-data network meta-analysis:13 studies, 348 participants | No significant POL vs PYR difference for VO₂max or time trial using time in heart-rate zones |
The 2014 result was from laboratory testing, not a race, in already trained people. The 2024 synthesis found the VO₂peak advantage mainly in interventions under twelve weeks and highly trained athletes. A VO₂peak gain does not automatically equal the same race-performance gain.23
The 2025 synthesis included 296 men and 52 women, 150 recreational and 198 competitive athletes. Level-based analysis suggested different responses: potentially greater aerobic benefits from POL in competitive athletes and PYR in recreational athletes. This subgroup result is not an algorithm that chooses a plan from the label “recreational”. The abstract found no significant overall difference between POL and the other models.4
The two meta-analyses are not simple replications: included studies, analyses and classification differ. A winner cannot be selected by adding one experiment’s percentages to pooled estimates.
Use it to read training
- Choose a zone scale and counting method before calculating shares.
- Look beyond one week: season phase and race specificity can change distribution.
- Five sessions do not create an obligation for one or two hard sessions: recovery, experience and volume remain individual decisions.
- Do not label a long session easy merely because it is long. Actual intensity and specific segments matter.
- Count running and cycling separately before aggregating multisport training: studies use different protocols and populations.
Sources
- Seiler S. What is best practice for training intensity and duration distribution in endurance athletes?,2010. PubMed, abstract.
- Stöggl T,Sperlich B. Polarized training has greater impact on key endurance variables than threshold, high intensity, or high volume training,2014. Frontiers, full text.
- Oliveira PS et al. Comparison of Polarized Versus Other Types of Endurance Training Intensity Distribution on Athletes’ Endurance Performance: A Systematic Review with Meta-analysis,2024. PMC.
- Rosenblat MA et al. Which Training Intensity Distribution Intervention will Produce the Greatest Improvements in Maximal Oxygen Uptake and Time-Trial Performance in Endurance Athletes?,2025. Sports Medicine, abstract.
Sources consulted on 29 October 2026. Illustrative counting examples.