Polarized, Pyramidal and Threshold
Introduction
Once you know where your thresholds are, the programming question follows: how much time do you spend in each territory? That's called intensity distribution, and it's where endurance training gets decided.
The popular answer — the polarized model, 80 % easy and 20 % hard — comes from observing elite athletes. That's worth knowing before copying it, because you probably don't train 15 hours a week.
The 80/20 isn't a physiological law. It's what people who train many hours do, and it works because they train many hours.
The three models
Polarized. Roughly 80 % of the time below the first threshold and 20 % above the second, with very little in between.
Pyramidal. Most of it easy, a portion at threshold, and a smaller portion very hard. The distribution decreases as intensity rises.
Threshold. A good share of volume concentrated in the moderate zone, around the second threshold.
Where the 80/20 came from
Stephen Seiler observed how elite endurance athletes — skiers, rowers, runners — actually trained, and found a consistent pattern: the vast majority of their sessions were clearly easy, and a minority clearly hard. The middle zone was sparsely populated.
The interpretation is reasonable: when you train 10 to 20 hours a week, the only way to accumulate that volume without breaking down is for most of it to be gentle. And for the hard sessions to deliver, you have to arrive at them rested.
In other words, the 80/20 solves the problem of accumulating high volume. That's its context.
What the evidence says comparing models
Studies comparing distributions find mixed results. Some meta-analyses show a slight advantage for polarized over threshold models; others find no clear differences between polarized and pyramidal.
In practice, elite athletes themselves tend to use pyramidal during base phases and something closer to polarized near competition, which suggests neither is superior in the abstract.
And there's a critique worth knowing: several authors have pointed out that the evidence favouring polarized training is weaker than its popularity suggests, and that much of its prestige comes from observational studies in elite athletes rather than controlled trials in ordinary people.
The nuance almost nobody mentions
Here's what matters for this book's reader.
If you train endurance three hours a week, devoting 80 % to very easy pace means two and a half hours of gentle jogging and half an hour of hard work. At that total volume, the polarized logic weakens considerably: the problem it solves — accumulating volume without breaking down — isn't your problem, because your volume is already low.
With limited hours, a distribution with a higher proportion of moderate and hard work is defensible, and it's what research on short-duration high-intensity training tends to show: with little time, intensity buys more adaptation per minute.
The practical rule that follows:
- Under 3 hours a week: a larger share of threshold work and intervals. You can approach 50/50.
- 3 to 6 hours: something intermediate, pyramidal in shape.
- Over 6 hours: there the 80/20 starts to make sense, because the volume demands protection.
What it looks like in practice
For someone with three or four endurance sessions a week and about four hours:
- Two easy, long sessions. Conversation possible throughout.
- One threshold session. Blocks of 8 to 15 minutes at a sustainable but uncomfortable pace, with short recoveries.
- One interval session. Sets of 2 to 5 minutes hard, with full recovery.
And a rule that applies to any distribution: the easy has to be genuinely easy. The most common error isn't choosing the wrong model, it's that gentle sessions end up moderate and hard sessions end up not hard. There everything collapses into the same thing.
Common mistakes
Copying the 80/20 at low volume. Applying a solution without having the problem.
Doing easy sessions too fast. The universal error, and it comes from ego more than ignorance.
Doing hard sessions too gently. If you could talk during your intervals, they weren't intervals.
Changing distribution every month. You need at least eight weeks to know whether something worked.
Counting by sessions rather than time. A hard 30-minute session and an easy 90-minute one aren't "one and one".
Conclusion
The 80/20 describes what athletes training many hours do, and its logic rests on that volume. With three or four hours a week, a distribution weighted more toward threshold and intervals is perfectly defensible.
Work out your own this week with the clock, not from memory: add up the minutes of your endurance sessions by territory. If you find nearly all of it lands in the middle — neither easy nor hard — that's your problem, and it's fixed by making the easy slower and the hard faster, without changing a single hour of your week.
References
- Seiler, S. (2010). What is best practice for training intensity and duration distribution in endurance athletes? International Journal of Sports Physiology and Performance, 5(3).
- Rosenblat, M. A., Perrotta, A. S., & Vicenzino, B. (2019). Polarized vs. threshold training intensity distribution on endurance sport performance: A systematic review and meta-analysis. Journal of Strength and Conditioning Research, 33(12).
- Stöggl, T. L., & Sperlich, B. (2015). The training intensity distribution among well-trained and elite endurance athletes. Frontiers in Physiology, 6.
- Burnley, M., Bearden, S. E., & Jones, A. M. (2022). Polarized training is not optimal for endurance athletes. Medicine & Science in Sports & Exercise, 54(6).
- Gibala, M. J., & Little, J. P. (2020). Physiological basis of brief vigorous exercise to improve health. The Journal of Physiology, 598(1).