Polarized Training

Shake it up a bit.


There are two prevailing approaches to training intensity distribution19,20:

  1. The threshold model: training is conducted predominantly within “zone 2”.
  2. The polarized model: the majority (~80%) of training is performed well below the lactate threshold (“zone 1”), with the remaining time allocated to intensities nearer VO2max (> “zone 3”)21.

a three zone model of exercise intensity categorisation A three-intensity-zone model delimited by ventilatory thresholds (Ref. 21). Put simply, an athlete would be able to talk comfortably in “zone 1”, less comfortably in “zone 2”, and would be in a mute world of hurt in “zone 3” (Ref. 9).

the lactate threshold training model versus the polarized training model Ref. 19

An inadvertent tendency towards the threshold model is often observed among amateur athletes6, partly due to training “too hard on easy days, and too easy on hard days6. Moreover, many high-level coaches also espouse the threshold training model1, perhaps reflecting the former currency of this approach (and/or evidence based on untrained subjects)4,19. However, in the last decade the polarized model has gained popularity16.

Support for the Polarised Model

  • Billat and colleagues1 studied a sample of elite Kenyan distance runners, all belonging to the Gusii tribe, roughly 2 weeks out from their competition phase. Based on their training logs, athletes were classified as either having trained to a “tempo” model or to a “speed” model (analogous to threshold and polarized training, respectively). Although this was a purely cross-sectional study (i.e. runners weren’t allocated to either condition), it is interesting that those in the “speed” training group were on average 2% (40s) faster over 10km than those in the “tempo” group, also faring better in several predictive measures of performance.
  • Schumacher & Mueller18 documented the training of the German 4 km team pursuit squad prior to the 2000 Olympic games (where they set a new world record). Outside of competition (e.g. stage races), ~95% of their training was completed at low intensities, with ~5% completed in zone 3. Perhaps more surprising, riders only trained on the track 20 days out from the games themselves.
  • Yu and colleagues23 modified the training schedule of top-level sprint speed skaters from their existing threshold model towards a more polarized model. The result was a significant (~3%) improvement in performance compared to their previous season(s).
  • Esteve-Lanao and colleagues4 experimentally allocated 20 regional/national level Spanish distance runners to one of two training programs (lasting 5 months): a threshold model, with a percentage time in zones 1/2/3 of 65/25/10; and a polarized model, yielding an intensity distribution of 80/10/10. Polarized training led to a ~2% greater improvement in 10km cross-country time compared to threshold training.
  • Stoggl & Sperlich22 took a group (N=41) of Austrian endurance athletes, from various sports (e.g. cross-country skiing, running, cycling, triathlon), and assigned them to one of four training programs. Lasting 9 weeks, these programs emphasized either high-volume, threshold, or HIIT training, and also included a combination of high-volume and HIIT training (i.e. polarized training). Those in the polarized training group displayed the largest improvements across all measures.
  • Neil and colleagues14 recruited 11 amateur cyclists to compare 6 weeks of polarized training (80/0/20 zone distribution) with 6 weeks of threshold training (55/45/0). The cyclists completed both training protocols, in a random order, with a 4 week detraining period in between. Despite the threshold training resulting in ~16% greater training volume, and an ~18% greater “load”, all performance measures increased to a greater extent in the polarized condition.


Optimization of endurance training requires that adaptive signaling be maximized and stress responses be minimized. It is on this basis that the polarized training model seems to be reasoned.

The first ventilatory threshold, demarcating the upper boundary of zone 1, seems to also represent a putative “stress” threshold20. This study20 found that running for 30mins at ~85% VO2max elicited a similar stress response to an interval session of 6x 3 minutes at 100% VO2max. The question then arises: if tempo and interval sessions are similarly “stressful”, are they similarly effective? The answer: probably not, particularly in well-trained athletes. This can be explained by a crude dichotomy of adaptive pathways, which distinguishes between those related to prolonged muscle activity and those related to energy crisis11. Thus, when the aim is to stimulate pathways related to energy crisis (i.e. high-intensity training), benefits will be commensurate with intensity10,15,17.

Commonplace in endurance lexicon is the term “junk miles”. Although it depends on who you are speaking to, “junk miles” is generally used to belittle high-volume LSD training. In light of recent understanding, however, it might be more appropriate to describe moderately high intensity training as, in fact, “junk miles”4. This form of training is neither potent enough to stimulate meaningful adaptation12, nor easy enough to permit high volumes and/or rapid recovery4. This shortfall is exacerbated when threshold training becomes the mainstay of a program— known as monotonic training—likely causing excessive fatigue and suboptimal adaptation7.

While HIIT training is an effective training device, athletes will be well aware of its limited application. In order to maintain the necessary intensity, HIIT sessions can only be scheduled so frequently, otherwise fatigue will accumulate and session quality will suffer4,5,22—termed a “ceiling effect”21. This is especially true if intervening sessions are performed at too high an intensity3. In addition to not hindering recovery, perhaps the greatest advantage of LSD training is that it has a far higher “ceiling”. Athlete’s can feasibly tolerate very high volumes of LSD training20, and reap benefits in near proportion5.

relationship between low intensity training volume and 10 kilometre running performance Ref. 5

Some have also speculated that polarized training might be more compatible with our evolutionary heritage13. Indeed, the hunter-gather lifestyle, which persisted until relatively recently (as far as our genomes are concerned), would have been characterized by prolonged “tracking persistence” interspersed with high-intensity bouts13. Thus, it seems only logical that we should benefit more from this pattern of activity.

Practical Considerations

The above section put forward the case for a polarized organization of training. However, there are important qualifications that need to be addressed:

  • The concept of training polarization needs to be incorporated as part of a wider, periodized training plan.
  • Excessive threshold training is certainly ill-advised. However, there is value in threshold training from a more psychological perspective. Sustained, high-intensity efforts can forge a very race-specific mental toughness, so should not be entirely neglected. It is interesting to note that one study2 of national level Portuguese & French marathon runners found only ~4% of these athletes’ training time was spent at marathon pace.
  • No discussion of human physiology would be complete without the ubiquitous caveat of individual variation. Despite the strong rationale behind polarized training, not everyone will respond optimally to it. Training must be tailored to the innate training preferences of any given athlete8,19.

Take Home Messages

High-volume, low-intensity (zone 1) training, supplemented by high-intensity (zone 3) work11, should form the backbone of any good endurance-training program. Threshold sessions (zone 2) are valuable, but are best used sparingly. The exact percentage distribution of these zones should be adjusted according to a periodized plan, and ought to reflect an athletes own training preferences.

Ultimately it boils down to this: don’t get caught in the “moderate middle”.


  1. Billat V, Lepretre PM, Heugas AM. Training and bioenergetic characteristics in elite male and female Kenyan runners. Med. Sci. Sports Exerc., 2003. doi: 10.1249/01.MSS.0000053556.59992.A9.
  2. Billat VL, Demarle A, Slawinski J, Paiva M. Physical and training characteristics of top-class marathon runners. Med. Sci. Sports Exerc. 33: 2089–2097, 2001.
  3. Bruin G, Kuipers H, Keizer HA. Adaptation and overtraining in horses subjected to increasing training loads. J. Appl. Physiol. 76: 1908–1913, 1985.
  4. Esteve-Lanao J, Foster C, Seiler S, Lucia A. Impact of training intensity distribution on performance in endurance athletes. J. Strength Cond. Res. 21: 943–949, 2007.
  5. Esteve-Lanao J, Juan AS, Earnest CP, Foster C. How do endurance runners actually train? Relationship with competition performance. Med. Sci. Sports Exerc., 2005. doi: 10.1249/01.MSS.0000155393.78744.86.
  6. Foster C, Heinmann KM, Esten PL, Brice G, Porcari JP. Differences in Perceptions of Training Coaches and Athletes. South African Journal of Sports Medicine 8: 3–7, 2001.
  7. Foster C. Monitoring training in athletes with reference to overtraining syndrome. Med. Sci. Sports Exerc. 30: 1164–1168, 1998.
  8. Gaskill SE, Serfass RC, Bacharach DW, Kelly JM. Responses to training in cross-country skiers. Med. Sci. Sports Exerc. 31: 1211–1217, 1999.
  9. Jeans EA, Foster C, Porcari JP, Gibson M. Translation of exercise testing to exercise prescription using the talk test. J. Strength Cond. Res. 25: 590–596, 2011.
  10. Laursen PB, Shing CM, Peake JM, Coombes JS, Jenkins DG. Influence of high-intensity interval training on adaptations in well-trained cyclists. J. Strength Cond. Res. 19: 527–533, 2005.
  11. Laursen PB. Training for intense exercise performance: high-intensity or high-volume training? Scand. J. Med. Sci. Sports 20: 1–10, 2010.
  12. Londeree BR. Effect of training on lactate/ventilatory thresholds: a meta-analysis. Med. Sci. Sports Exerc. 29: 837–843, 1997.
  13. Lucia A, Boullosa DA, Martin DT, Nakamura FY, Hawley JA, Ruiz JR, Esteve-Lanao J, Seiler KS. Letters to the Editor. Int. J. Sports Physiol. Perform. 5: 431–436, 2010.
  14. Neal CM, Hunter AM, Brennan L, O’Sullivan A, Hamilton DL, DeVito G, Galloway SDR. 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. 114: 461–471, 2013.
  15. Niklas P, Li W, Jens W, Michail T, Kent S. Mitochondrial gene expression in elite cyclists: effects of high-intensity interval exercise. Eur. J. Appl. Physiol. 110: 597–606, 2010.
  16. Orie J, Hofman N, de Koning JJ, Foster C. Thirty-eight years of training distribution in olympic speed skaters. Int. J. Sports Physiol. Perform. 9: 93–99, 2014.
  17. Ronnestad BR, Hansen J, Vegge G, Tonnessen E, Slettalokken G. Short intervals induce superior training adaptations compared with long intervals in cyclists — An effort-matched approach. Scand. J. Med. Sci. Sports, 2014. doi: 10.1111/sms.12165.
  18. Schumacher YO, Mueller P. The 4000-m team pursuit cycling world record: theoretical and practical aspects. Med. Sci. Sports Exerc. 34: 1029–1036, 2002.
  19. Seiler KS, Kjerland GO. Quantifying training intensity distribution in elite endurance athletes: is there evidence for an “optimal” distribution? Scand. J. Med. Sci. Sports 16: 49–56, 2006.
  20. Seiler S, Haugen O, Kuffel E. Autonomic recovery after exercise in trained athletes: intensity and duration effects. Med. Sci. Sports Exerc., 2007. doi: 10.1249/mss.0b013e318060f17d.
  21. Seiler S. What is best practice for training intensity and duration distribution in endurance athletes? Int. J. Sports Physiol. Perfom. 5: 276–291, 2010.
  22. Stoggl 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/abstract.
  23. Yu H, Chen X, Zhu W, Cao C. A Quasi-Experimental Study of Chinese Top-Level Speed Skaters’ Training Load: Threshold Versus Polarized Model. *Int. J. Sports Physiol. Perform.** 7: 103–112, 2012.