Polarised training and 80/20
Evidence: contested
“Keep most running easy” is well supported. That the specific polarised 80/20 split is optimal is contested. “Zone 2” also means different things in different systems, a clash worth keeping in mind.
Polarised training describes a distribution in which most running is easy, below the first lactate threshold, a meaningful minority is genuinely hard, above the second threshold, and little sits in the middle “threshold” zone. Seiler first reported the pattern in a small sample of 11 junior cross-country skiers, who clustered around 75 to 80% easy and 15 to 20% hard, a roughly U-shaped distribution (Seiler & Kjerland 2006); wider observational work in elite endurance athletes since has broadly echoed it.
The 80/20 figure depends on how you count it
The split can be measured two ways, and they do not give the same number. Counting the proportion of training time in each zone is not the same as counting the proportion of sessions, and Seiler’s own work has used both, which is part of why the figure is quoted loosely (Seiler & Kjerland 2006). A runner can do a minority of sessions hard yet still spend about 80% of total time easy, because easy runs are longer. When comparing one plan with another, check which basis the 80/20 is measured on before treating the numbers as equivalent.
A warning about “Zone 2”
The same word, “zone two”, means opposite things in the two conversations runners overhear, and the clash causes real confusion. In the three-zone model that underpins polarised training, Zone 2 is the middle, between the two lactate thresholds, and polarised training deliberately minimises it. In the popular “Zone 2 training” trend, and in the five-zone scheme on most watches, “Zone 2” means easy aerobic running, low and comfortable, which polarised training wants lots of. So “do plenty of Zone 2” (the popular sense) and “stay out of Zone 2” (the polarised sense) are both correct advice pointing at different intensities. The underlying message of both, keep most running genuinely easy, is the same and is well supported; the label is the trap. When a plan or a device says “Zone 2”, check which model it means before acting on it. See training philosophies for the wider definitional mess.
What the evidence supports, and what it does not
The claim that most training should be easy is well supported, and overlaps with the volume-versus-intensity evidence. The stronger claim that the specifically polarised 80/20 split is optimal is contested:
- A frequently cited trial found polarised training produced the largest gains in VO₂peak and time to exhaustion, but it is also the most criticised, with small groups and baseline imbalances (Stöggl & Sperlich 2014).
- When high-intensity work is held equal, more easy volume improved 10 km cross-country time more than less, which reads as support for a pyramidal lean rather than a strict polarised one (Esteve-Lanao et al. 2007).
- Highly trained runners in practice often follow a pyramidal distribution, with some genuine threshold work, especially away from competition (Casado et al. 2022).
- A 2024 meta-analysis of 17 studies found no evidence that polarised training is superior to other distributions for VO₂peak, time-trial, time to exhaustion, or speed at threshold (Silva Oliveira et al. 2024).
The expert literature treats this as a live debate rather than a settled question (Burnley, Bearden & Jones 2022). The honest position: keep most running easy because that is well evidenced; treat the exact hard-easy ratio as something to tune to the individual, not a law.