Training models

Polarised, pyramidal and threshold training

Training intensity distribution is how your training time splits between easy, moderate and hard. Polarised training puts almost everything at the two ends, pyramidal tapers from lots of easy to a little hard, and threshold-focused training builds around controlled work in the middle.

Whichever model they follow, most successful endurance athletes spend the large majority of their time training easy. The models disagree mainly about what to do with the rest.

By Nick Buckley-Mills, Team GBR age-group triathlete Reviewed 15 October 2026 7 min read

At a glance

Model	Rough shape	Associated with	Often suits
Polarised	Mostly easy, little moderate, a meaningful slice hard	Stephen Seiler's research on elite endurance athletes	Athletes with enough volume for easy work to add up
Pyramidal	Mostly easy, less moderate, least hard	The pattern most often recorded in elite athletes	A broad default across sports and levels
Thresholdalso "Norwegian"	High easy volume plus several tightly controlled threshold sessions a week	Marius Bakken in running; Olav Aleksander Bu's group in triathlon	High-volume athletes who can hold intensity precisely
Sweet Spot	More time just below threshold, less pure easy volume	Power-based cycling plans for limited hours	Cyclists and triathletes short on time

What are "easy", "moderate" and "hard"?

Research usually splits intensity into three zones, set by two physiological thresholds. Below the first threshold, effort is easy and sustainable for hours. Above the second, it is hard and sustainable for minutes. Between them is the moderate middle, sometimes called the grey zone. Those thresholds are the same turn points a lactate test looks for.

These three research zones are not the same as the five or seven zones in most training apps. "Zone 2" in a five-zone app usually sits inside research Zone 1. Mixing the two systems is behind a lot of confusion about what polarised training actually asks for. See how training zones work.

How do the models differ?

Polarised

Keep easy days genuinely easy and hard days genuinely hard, with little in between. Stephen Seiler's research on endurance athletes, including a study of elite junior cross-country skiers, found roughly three quarters of sessions were easy, a small share moderate, and the rest hard. The idea spread to amateurs as "80/20", popularised by Matt Fitzgerald.

Pyramidal

Most training easy, a steady layer of moderate work, a small amount of hard. When researchers look back at what well-trained and elite athletes actually did across a season, this is the shape that shows up most. Many shift towards a more polarised pattern as competition approaches.

Threshold, or the "Norwegian" model

A large base of easy volume, plus three or four threshold sessions a week, often guided by blood lactate. The approach grew out of Norwegian middle- and long-distance running and is associated in triathlon with coach Olav Aleksander Bu's athletes. The case for it is that threshold work, kept carefully under control, produces less fatigue than very hard intervals, so you can do more of it.

Sweet Spot

Spend more time just under threshold to get a lot of training effect from limited hours. It is common in time-limited cycling plans. The trade-off is fatigue: done week after week, it leaves less room for both real recovery and real intensity.

What does the research say?

No model clearly beats the others for race performance. A 2014 study by Thomas Stöggl and Billy Sperlich found nine weeks of polarised training improved key endurance measures more than threshold, high-intensity or high-volume training in well-trained athletes. A 2024 meta-analysis pooling 17 studies found a small overall edge for polarised training on VO2peak, clearest in interventions shorter than 12 weeks and in highly trained athletes. It found no clear difference in time-trial performance.

Most of these studies are short, small, and run on athletes who aren't juggling a full-time job. Their results tell you more about what is possible than about what will work for you.

Why does the way you count matter?

The same week can look polarised or pyramidal depending on how you count it. Count time in zone, and a 5 × 4-minute VO2 Max session is mostly easy: warm-up, recoveries and cool-down. Count by the purpose of each session, and it is one hard day. A 2015 review by Stöggl and Sperlich showed that heart rate, lactate, perceived effort and session goal can classify the same training differently, which partly explains why studies disagree.

Before you decide whether your training is "polarised", decide how you are measuring it, and stick to that.

What changes for triathletes?

You can, and often should, run a different distribution in each sport.

Hard running costs the most. Many triathletes keep the run more polarised and put more moderate work on the bike, where it is cheaper to recover from.
Long-course racing happens in the middle. Half and full Ironman pace sits in or near the moderate zone for most athletes. Race-specific work is grey-zone work by definition, so a strictly polarised plan has to make room for it.
Swimming follows its own rules. Pool sessions mix technique, short fast repeats and steady aerobic sets, and they rarely fit neatly into a three-zone split.
Elite triathletes vary too. A published case study of two world-class short-distance triathletes described a polarised distribution, while the Norwegian group is known for heavy threshold work. Both produced world-class results.

Should you pick a model?

Use the models to understand your training, not to define it. They are useful lenses, and they turn into dogma remarkably fast.

What they get right

They make you look at distribution at all. Many amateurs never check, and end up doing almost everything at the same moderate effort.
They agree on the big thing. Every model builds on a large base of easy training. That consensus is more useful than any of the disagreements.
They give you a vocabulary to describe and adjust your week.

Where they fall short

They describe, they don't prescribe. Most come from looking back at what successful athletes did. That isn't the same as proof the pattern caused the success, or that it will work for you.
They were built on big training volumes. Eighty percent easy means something different at 25 hours a week than at 8. At low volume, it is less clear the same split gives enough stimulus.
They flatten the athlete. A percentage split says nothing about your history, your limiter, your race, your job or how you respond. Two athletes on the "same" model can need very different weeks.
They are fragile to measurement. Change how you count intensity and you can change which model you appear to be following.
They become identities. "I'm polarised" can turn a sensible principle into a reason not to adapt when the evidence in front of you says otherwise.

How to use them anyway

Start from your race, your hours and your history, and pick the shape that fits them. Review your distribution over four to eight weeks rather than week by week. If you're stalling, change one thing about the shape and watch what happens. The label never trains you. The sessions do.

Common questions

Is 80/20 the same as polarised training?

Not quite. 80/20 refers to roughly 80 percent of training being easy and 20 percent moderate or hard. Polarised training goes further and keeps the hard share mostly at high intensity, with very little in the moderate middle.

Is Zone 2 training the same as polarised training?

No. Zone 2 in most five-zone apps describes easy aerobic work, which is the base of every model. Polarised training is about how the rest of your time is split, not just about doing easy training.

Can age-group triathletes use the Norwegian method?

Parts of it, with care. The model relies on high easy volume and tightly controlled threshold sessions, often checked with blood lactate. Without the volume or the control, threshold sessions easily drift too hard and become exhausting rather than productive.

Which model is best for Ironman?

No model is proven best. Long-course race pace sits near the moderate middle, so whichever model you follow needs room for race-specific work at that intensity, on top of a large base of easy training.

Sources

Seiler KS, Kjerland GØ. Quantifying training intensity distribution in elite endurance athletes: is there evidence for an "optimal" distribution? Scandinavian Journal of Medicine & Science in Sports. 2006;16(1):49–56.
Seiler S. What is best practice for training intensity and duration distribution in endurance athletes? International Journal of Sports Physiology and Performance. 2010;5(3):276–291.
Stöggl T, Sperlich B. Polarized training has greater impact on key endurance variables than threshold, high intensity, or high volume training. Frontiers in Physiology. 2014;5:33.
Stöggl TL, Sperlich B. The training intensity distribution among well-trained and elite endurance athletes. Frontiers in Physiology. 2015;6:295.
Silva Oliveira P, Boppre G, Fonseca H. Comparison of polarized versus other types of endurance training intensity distribution on athletes' endurance performance: a systematic review with meta-analysis. Sports Medicine. 2024;54(8):2071–2095.
Casado A, Foster C, Bakken M, Tjelta LI. Does lactate-guided threshold interval training within a high-volume low-intensity approach represent the "next step" in the evolution of distance running training? International Journal of Environmental Research and Public Health. 2023;20(5):3782.
Cejuela R, Selles-Perez S. Training characteristics and performance of two male elite short-distance triathletes: from junior to "world-class". Scandinavian Journal of Medicine & Science in Sports. 2023;33(12):2444–2456.
Fitzgerald M. 80/20 Running. New American Library, 2014.