Polarised vs Pyramidal Training: What the Evidence Says

Short answer: both work. Polarised training is mostly easy, with very little in the middle and a small amount of hard. Pyramidal training is mostly easy, with less in the middle and even less hard. They share the feature that matters most: a large majority of time at low intensity. The most recent analysis pooling individual athletes' data found no clear difference between the two for VO2max (the most oxygen your body can use) or race-like time trials (1).

For a time-limited amateur, the practical message is to protect the easy days and not let most of your training drift into the "comfortably hard" middle. Plenty of easy volume is the most consistent finding: it is what successful endurance athletes do, though that comes mainly from observing high-volume athletes, not from trials in time-limited amateurs. The evidence against too much middle work is weaker, mostly from reviews of small studies in runners (2, 3), but it points the same way.

What should a time-limited amateur do?

The recommendation for most people training 5–12 hours a week: keep about three quarters or more of your training time easy, and do two quality sessions a week, one hard and one steady or threshold. That is a pyramidal week. In the final weeks before a short, fast race, you can shift towards polarised (less middle, a little more hard).

What matters most depends on your experience:

• Newer athletes: the first priority is simple. Make easy sessions genuinely easy and add one harder session a week. The distribution question is a refinement for later.
• Experienced athletes: look at where your time actually goes. If many "easy" runs or rides drift into the middle, fix that first. Then decide on purpose how much middle work your race needs.

Results depend on starting fitness and on how many hours you have. Most studies used already-trained athletes, mostly men, over a few weeks to a few months (1, 4). Treat the advice below as a direction to test on yourself, not a prescription.

What do "polarised" and "pyramidal" actually mean?

Researchers usually split intensity into three bands, separated by your two lactate thresholds: the point where lactate (produced by working muscle and used as fuel elsewhere) first rises above resting levels, and the highest effort at which it can still hold steady (5). This article calls them Easy, Middle and Hard, because research papers call them zones 1–3, which clash with the five or more zones on most watches.

• Easy: below your first threshold. Conversational. You could go for hours.
• Middle: between the two thresholds. Steady, "comfortably hard", speaking in short sentences. Tempo runs and sweet-spot rides sit here, as does much of marathon or half-Ironman race pace for faster amateurs.
• Hard: above your second threshold. Intervals where you can only manage a few words.

Finding your bands without a lab. As a rough guide:

• Easy: below about three quarters of FTP on the bike (FTP is the highest power you can hold for roughly an hour), or below roughly 89% of your lactate threshold heart rate (LTHR), the top of Friel's endurance zone (6). The two won't always agree; on the bike, use power when you have it. You can estimate LTHR with Friel's 30-minute solo test (warm up well, and skip it if you have a heart condition or are new to hard efforts): your average heart rate over the last 20 minutes. If your watch zones are set from LTHR, Easy is roughly zones 1–2. Default zones based on maximum heart rate often put the first threshold in zone 3, so check how your zones are set before counting time in each.
• Middle: from there up to FTP, LTHR or threshold pace.
• Hard: anything above them.

No numbers? Use the talk test: full conversation is Easy, short sentences is Middle, a few words is Hard. Garmin Connect, Strava and similar apps show your time in each zone. Count easy days by heart rate, but count interval sessions by power or pace, because heart rate lags behind short efforts and makes them look easier than they were.

The names describe the shape of time in each band:

• Pyramidal: most time Easy, less Middle, least Hard.
• Polarised: most time Easy, very little Middle, more Hard than Middle.
• Threshold: a large share Middle, with less easy work. Most coaches advise against this as a default, though the trial evidence against it is weaker than the case for easy volume.

Two illustrations by time: a polarised week might be 80% Easy, 5% Middle and 15% Hard; a pyramidal week 75% / 20% / 5%. Both put most of the time in Easy. The difference is where the remaining 20–25% goes.

What do the best athletes do?

Mostly easy training. Studies of nationally and internationally competitive athletes find that about 80% of their sessions are done at low intensity (7), and world-class distance runners do 80% or more of their running volume at low intensity throughout the year (8). Most are not strictly polarised: reviews describe mainly pyramidal distributions, with some athletes shifting towards polarised in certain phases (9). In a review of highly trained distance runners, marathoners leaned pyramidal and 1500-metre runners polarised, and the authors recommend shifting towards polarised in the competition period (5).

Elite athletes also train 10 to 14 times a week (7, 8), so their easy share is enormous in absolute hours. An amateur doing 7 hours with the same percentages is doing a very different week.

How to use this in a real training week

Here is an example pyramidal week of about 7 hours for a triathlete in the middle of a training block:

• Monday: rest.
• Tuesday (Hard), run, 60 minutes: 15–20 minutes easy with a few short pick-ups, then 5 × 4 minutes hard with 3 minutes easy jogging between, then 10 minutes easy. Hard means you could only say a few words, roughly your 5K effort. Stop the set if you can't hold the pace of the first repeat.
• Wednesday (Easy), swim, 45 minutes: easy, with some technique work.
• Thursday (Middle), bike, 75 minutes: 2 × 15 minutes steady or at threshold with 5 minutes easy between, inside an easy ride. Comfortably hard: short sentences, getting harder towards threshold.
• Friday (Easy), swim, 45 minutes.
• Saturday (Easy), long ride, about 2¼ hours.
• Sunday (Easy), run, 70 minutes.

By time, this comes out at roughly 88% Easy, 7% Middle and 5% Hard, with warm-ups, recoveries and cool-downs counted as easy. It is still pyramidal, just with a bigger easy share than the illustrations above, which is fine. By sessions, though, two out of six is about a third. Elite athletes do roughly one hard session in five, with their two or three hard days spread across many more sessions (7). Few weeks match 80/20 by both sessions and time. Time is easier to read off your watch; sessions better reflect how much hard work you have to recover from. Check both, and if they disagree, let how you recover decide.

Runners: make every session a run. Easy days of 40–50 minutes, Thursday as a 60-minute run with the two 15-minute efforts, and a long run of about 90 minutes instead of the long ride. That keeps the long run at roughly a quarter of weekly running, as Daniels suggests (10).

What easy should feel like: you can hold a full conversation, and you'd happily do it again tomorrow. If easy days leave you tired, they may be too hard or have too many hard days around them, but also look at total volume, sleep, fuelling and stress.

When a quality session goes badly: cut it short, finish the time easy, and move on. If it happens two or three times in a row, look at sleep, stress and whether your easy days have crept into the middle.

Training for a long race? For faster amateurs racing a marathon or 70.3, much of race pace falls in the Middle band, so some Middle work is specific to the race, not wasted. For a full Ironman, most amateurs should pace at or just under the first threshold, so race-pace work sits at the top of Easy or the bottom of Middle. To see where your own race pace falls, compare your race heart rate or power with your easy ceiling.

A simple plan for the season

This is a common coaching structure, built partly on how elite runners organise their season (5, 8). It has not been tested as a whole in amateurs, so treat it as a sensible starting point:

1. Base phase: pyramidal. Mostly easy, one steady or threshold session a week, and short, controlled hard efforts only occasionally.
2. Build phase: still pyramidal. Two quality sessions a week, one Middle and one Hard.
3. Final 4–8 weeks: make the quality sessions race-specific (see "Race-specific final weeks" below). Elite runners increase race-pace work as their main competition approaches and start tapering 7–10 days before it (8).
4. Every 3–4 weeks: take a lighter week, and check where your time actually went, not where you planned it to go.

Other options

Race-specific final weeks. For short, fast races (5K, sprint triathlon), shift towards more Hard and less Middle, close to polarised. For a 10K or Olympic-distance triathlon, which many amateurs race close to their threshold (a 10K under about an hour at or a little above it), keep threshold work as one of the two sessions and add some Hard. For long races, stay pyramidal with race-pace work.

Strictly polarised all year. A reasonable choice, especially for well-trained athletes racing shorter events. Keep Hard to one or two sessions a week, and don't let it grow because "hard" feels productive.

Very low hours (under 5 a week). As a rule of thumb, percentages stop being useful here, so think in sessions. Newer athletes: one quality session plus easy volume. Experienced athletes who recover well can often manage two short ones, as long as everything else stays genuinely easy. Drop back to one if quality sessions start failing.

Threshold-heavy training. This is common when every session becomes a tempo effort. Reviews of runners suggested that pyramidal and polarised distributions tend to do better than threshold-dominant ones (2, 3), though they drew on small studies and the 2025 pooled analysis found no significant difference (1). Threshold training is also used by some of the world's best marathon runners (2). Those athletes do controlled threshold work on top of very high easy volume (8). So do followers of the Norwegian method, a practitioner's approach brought to triathlon by coach Olav Aleksander Bu. For 7 hours a week, treat a week that is mostly Middle as a warning sign, not a plan.

The science behind it

Optional reading, for anyone who wants the studies.

The early trials favoured polarised. In a nine-week study of 48 well-trained runners, cyclists, triathletes and cross-country skiers, the polarised group improved VO2peak by 11.7%, more than high-volume, threshold or high-intensity groups (11). In a crossover study of 12 trained male cyclists, six weeks of polarised training improved peak power and lactate threshold more than a threshold programme, even though the polarised block involved fewer hours, 6.4 against 7.5 a week; the 40-km time trial improved with both (12). The cyclist study compared polarised with threshold training only. The nine-week study's high-volume group came closest to a pyramidal week, so the early evidence on polarised against pyramidal rests on one small trial.

Recreational runners showed a smaller difference. In a 10-week study of 30 recreational runners, 10K times improved by 5.0% with polarised and 3.6% with between-thresholds training, a difference that was not statistically significant (13). Only when the authors looked at the runners who most closely followed each plan did polarised come out ahead.

The meta-analyses are mixed. An early meta-analysis found a moderate effect favouring polarised over threshold training for time trials, but it pooled only three studies of modest quality (14). A larger 2024 meta-analysis of 17 studies and 437 people found polarised slightly better for VO2peak, an effect seen only in interventions shorter than 12 weeks and in highly trained athletes. Time trials and threshold power or speed improved similarly with polarised and other distributions (4).

The 2025 network meta-analysis of individual participant data, from 13 studies and 348 athletes, found no difference between polarised and pyramidal for VO2max or time-trial performance, and no significant difference between polarised and any other distribution. A subgroup analysis suggested that, for VO2max, competitive athletes may improve more with polarised and recreational (but already trained) athletes more with pyramidal (1). Only 52 of the 348 participants were women.

Why the studies disagree. Different ways of counting zones, by heart rate, lactate, pace or perceived effort, give different distributions for the same training (3). Most trials last weeks, not seasons, and compare small groups. Comparators vary: "threshold" in one study is not "pyramidal" in another. To make labels consistent, one group proposed a polarisation index, a single number worked out from the three shares; above 2.00 counts as polarised (15). By that measure, the 80/5/15 illustration is polarised (about 2.4) and the 75/20/5 one is not (about 1.3).

Taken together, the evidence suggests that a large base of easy training is the common ground and that the difference between polarised and pyramidal is small for most people. Threshold-dominant training tends to come out worst in the smaller trials and runner reviews, though the 2025 individual-data analysis found no statistically significant difference (1).

References

1. Rosenblat MA, Watt JA, Arnold JI, et al. Which Training Intensity Distribution Intervention will Produce the Greatest Improvements in Maximal Oxygen Uptake and Time-Trial Performance in Endurance Athletes? A Systematic Review and Network Meta-analysis of Individual Participant Data. Sports Med. 2025. doi:10.1007/s40279-024-02149-3
2. Kenneally M, Casado A, Santos-Concejero J. The Effect of Periodization and Training Intensity Distribution on Middle- and Long-Distance Running Performance: A Systematic Review. Int J Sports Physiol Perform. 2018. doi:10.1123/ijspp.2017-0327
3. Campos Y, Casado A, Vieira JG, et al. Training-intensity Distribution on Middle- and Long-distance Runners: A Systematic Review. Int J Sports Med. 2022. doi:10.1055/a-1559-3623
4. 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 Med. 2024. doi:10.1007/s40279-024-02034-z
5. Casado A, González-Mohíno F, González-Ravé JM, et al. Training Periodization, Methods, Intensity Distribution, and Volume in Highly Trained and Elite Distance Runners: A Systematic Review. Int J Sports Physiol Perform. 2022. doi:10.1123/ijspp.2021-0435
6. Friel J. The Triathlete's Training Bible. 4th ed. VeloPress; 2016.
7. Seiler S. What is best practice for training intensity and duration distribution in endurance athletes?. Int J Sports Physiol Perform. 2010. doi:10.1123/ijspp.5.3.276
8. Haugen T, Sandbakk Ø, Seiler S, et al. The Training Characteristics of World-Class Distance Runners: An Integration of Scientific Literature and Results-Proven Practice. Sports Med Open. 2022. doi:10.1186/s40798-022-00438-7
9. Stöggl TL, Sperlich B. The training intensity distribution among well-trained and elite endurance athletes. Front Physiol. 2015. doi:10.3389/fphys.2015.00295
10. Daniels J. Daniels' Running Formula. 3rd ed. Human Kinetics; 2014.
11. Stöggl T, Sperlich B. Polarized training has greater impact on key endurance variables than threshold, high intensity, or high volume training. Front Physiol. 2014. doi:10.3389/fphys.2014.00033
12. Neal CM, Hunter AM, Brennan L, et al. Six weeks of a polarized training-intensity distribution leads to greater physiological and performance adaptations than a threshold model in trained cyclists. J Appl Physiol (1985). 2013. doi:10.1152/japplphysiol.00652.2012
13. Muñoz I, Seiler S, Bautista J, et al. Does polarized training improve performance in recreational runners?. Int J Sports Physiol Perform. 2014. doi:10.1123/ijspp.2012-0350
14. Rosenblat MA, Perrotta AS, Vicenzino B. Polarized vs. Threshold Training Intensity Distribution on Endurance Sport Performance: A Systematic Review and Meta-Analysis of Randomized Controlled Trials. J Strength Cond Res. 2019. doi:10.1519/jsc.0000000000002618
15. Treff G, Winkert K, Sareban M, et al. The Polarization-Index: A Simple Calculation to Distinguish Polarized From Non-polarized Training Intensity Distributions. Front Physiol. 2019. doi:10.3389/fphys.2019.00707