Polarized Training

Deeper Dive

Because of the heterogeneity in the studies I’m going to take a deeper look at two; one about the performance of recreational runners and another about club level cyclists. Both these groups feature trained, but sub-elite runners/cyclists. The subjects are in their mid-30s with the running group training 4-6 hours per week and the cyclists 7-8. This is a pretty representative sample of what endurance training looks like for most dedicated amateurs. And, both of these studies are very methodologically sound, so we can learn a lot from the results. Cyclists up first.

Six weeks of a polarized training-intensity distribution leads to greater physiological and performance adaptations than a threshold model in trained cyclists

This study was done on male cyclists recruited from two different cycling clubs, and compared the results of a six week polarized training plan with a six week threshold based training plan. In order to participate they had to have been consistently training for at least four years, training four to five times a week for the last 6 months, and complete a simulated 40-km Time Trial at an average power greater than 240 watts. Not elite, but experienced and competitive.

The subjects ran through a battery of physiological tests to establish their baseline (40-km TT performance, power at lactate threshold 1 and 2 (roughly 2.0 and 4.0mmol), maximal aerobic power, and time to exhaustion at 95% of maximal aerobic power) before beginning the study. The researchers implemented a very rigorous study design. It was conducted in a crossover fashion (each subject completed both training interventions and functioned as their own control) and had multiple 4-week periods where subjects only did low-intensity training to help minimize interference effects and bring subjects back to baseline (referred to as a “controlled detraining period”). All in all it took 28 weeks from start to finish. Check out the graphic below for a visual breakdown of the study:

In terms of training interventions the Polarized group performed 80% of their training as low-intensity endurance in zone 1, 20% as high intensity in zone 3, and no training in zone 2. The threshold group did 55% of their training as low-intensity endurance in zone 1, 45% as medium intensity in zone 2, and no training in zone 3. This is an extreme difference in intensity distribution to make sure that the two programs are effective contrasts. The Polarized group did intervals of 6x4min while the Threshold group did 60min at an intensity in between their lactate thresholds.

So, how did it turn out? When the results shook out both groups saw improvements across the board, but the Polarized group ended up having better results for every single metric. The largest differences between groups were seen at the intensity extremes, with the Polarized group increasing their maximal aerobic power (MAP), their time to exhaustion at 95% of MAP, and power output at the first lactate threshold (LT1). They also improved more for the 40km TT, and power output at the second lactate threshold, but these results were smaller and not statistically significant.

LT1 is the upper limit of Zone 1 training, and MAP is the upper end of Zone 3 training, so it is not shocking that a training plan focused on these two extremes caused more improvement in the corresponding variables. This adheres pretty well to the principle of specificity; those two qualities were targeted and they improved accordingly. Interestingly however, the Polarized group also saw more improvement than the Threshold group (though, not statistically significant) in both their power at LT2 and 40km TT performance. Those two values are very similar, and also close to the power output used during the Threshold interventions. That would make it seem like the threshold training would be more specific to those tests, however despite this, the subjects saw better results during polarized training.

Caveat: it was estimated that subjects had on average been following a 53/38/9 training plan that is VERY similar to the 57/43/0 threshold intervention. Could it be that the polarized plan wasn’t more effective, but simply it was a different stimulus than what they were used to and the results would have been reversed if the cyclists had already adapted to a polarized training plan? Additionally, the interventions were fairly short (6 weeks), so it is possible that during a longer cycle the threshold intervention would catch up to the polarized group. A lot of research on higher intensity intervals shows that they (zone 3 training in this example) cause rapid improvements that plateau after 6-8 weeks.

Even with that in mind, with the methodologically sound design it’s hard not to take away that the polarized group had better results across the board. Now, that does not mean that the researchers came up with the optimal distribution, and it does not mean that you should have absolutely no zone 2 work in a training plan, but it’s more evidence in favor of a polarized approach.

As for the other study.

Does Polarized Training Improve Performance in Recreational Runners?

This study had a similar aim; take a group of amateur, but experienced, athletes and see if a polarized training program or a threshold based training plan would have better performance outcomes. The subject group for this study was 32 recreational runners, 16 of the runners followed a 10 week polarized training plan (intensity distribution of ~75/5/20) while the other half followed a 10 week threshold plan (intensity distribution of ~45/35/20). Subjects trained with heart rate monitors to measure their time in each zone. Like with the previous study, both groups followed the same 8 week preparatory training before beginning their training intervention. The main performance metric was a 10k race, one was performed before the intervention and a second one performed afterwards.

The results found that both groups improved following the training interventions, with the polarized group improving a little bit more than the threshold group (on average the polarized group improved by 1:59 and threshold by 1:23). However, these results were not statistically significant (the differences are not extreme/consistent enough for researchers to rule out that the changes between groups were due to random chance). This could be for a few reasons. For one, it could be that the differences between groups were purely random. However, there are a few other possibilities. For one, unlike the previous study this did not employ a cross-over design; subjects simply did one of the interventions instead of one and then the other. Because of this you compare two different groups of subjects instead of one subject to itself, this is going to drastically lower the statistical power and make it harder to detect significant differences.

Additionally, in this study the subjects did not do as great a job of keeping their training in the right zones; some of the subjects in the threshold group When each intervention was further broken down into two groups of six that were most extreme in terms of polarization (average intensity distribution of 78/11/11) and threshold (average intensity distribution of 32/53/16) the differences in performance were also more extreme with the polarized group improving on average by 7% and the threshold group improving on average by 1.6%. While this difference is certainly larger it needs to be taken with a grain of salt. Breaking small groups down into even smaller groups for analysis can be fraught with difficulties. In this case two of the runners in the small threshold group actually performed worse on the follow up test (the only two runners in the study to do worse on the follow up 10k). Was this because their training intervention was less effective due to the intensity distribution, or did they just have poor races? With such small sample sizes one or two outliers can really skew the data.

Takeaways

So, what does this all mean? Looking at the big picture, the polarized training interventions tended to have slightly better results. The study on cyclists saw a more significant difference between groups. This study had a bigger difference in terms of intensity distribution, did a better job of ensuring subjects performed workouts in the correct zones, and had a stronger design (the cyclists also performed more training per week than the runners). However in both studies the more threshold oriented interventions also saw significant improvements, and the differences between groups was not huge. In both cases the studies were not very long, did not have very many participants, and did not show huge changes between groups (these three issues will plague almost all studies on training). Overall, this is more evidence in favor of polarized training, but it is certainly not definitive proof.

It is also important to note that both studies had reasonably sized standard deviations which indicates a fair amount of variability in response to training. The results did not show that all of the subjects in the polarized group improved more than all of the subjects in the Some athletes responded very well to, others not as well. I would be very interested in seeing the individual data in the cycling study since their subjects performed both interventions; did athletes who responded well to the threshold intervention not respond as well to the polarized one? And, what about vice versa?

While it is clear that more research on the topic needs to be done I am in general a proponent of a polarized intensity distribution for endurance. That does not mean it is a silver bullet, and it does not mean it is perfect for everyone; in training there are no one-size-fits-all solutions. However, the principles of polarized endurance training can be fairly easily applied by many people to improve the structure of their training.

Why Does This Work?

While there has been a lot of research on the polarization of athletes based on their training logs, and an increasing amount of experimental trials concerning polarized training, there has not been a lot of research on why it is more effective. The current hypothesis is that the main benefit of low-intensity endurance training is that athletes are able to recover from it much quicker than they are moderate intensity or high intensity (Laursen, 2010). This would mean they can train more often and with higher volume than if all of their training was harder. Researchers conducted a study to try and evaluate this hypothesis (Seiler, Haugen, and Kuffel, 2007).

They had highly trained runners complete four different workouts representing a few different intensity levels; low-intensity runs of 60 or 120 minutes, 30 minutes of moderate intensity running, or six sets of three minute intervals at high intensity. Following each workout researchers monitored the recovery of their autonomic nervous system via heart rate variability (HRV). Researchers found that after both the 60 and 120 minutes of low-intensity training the athletes returned to baseline values of HRV within 5 minutes. Moderate and high intensity training however both took roughly 60 minutes to return to baseline, with no significant difference between the two intensities. This indicates that despite moderate intensity training being less taxing on the athletes (as observed by having a significantly lower rating of perceived exertion), the recovery demands might be similar. So, it might have fewer benefits, but roughly the same drawbacks There are a few caveats to this study; HRV is not a complete measure of stress or recovery, the athletes only had recovery assessed following the session with no follow up the next day, only acute measures of recovery was looked at, and the study consisted of a small group of highly trained runners. However, it definitely lends some support to the hypothesis.

Another possible explanation comes from current evidence concerning exercise induced mitochondria biogenesis. Improving the amount and the efficiency of mitochondria is a key target of endurance training with fairly rapid changes being observed after short periods of training (Bishop, Granata, and Eynon, 2014). The specifics of mitochondria biogenesis goes beyond the scope of this article, but improvements in mitochondrial volume and efficiency can be achieved both by increasing intensity of training as well as increasing volume. However, based on the available research it seems that there is a strong correlation between higher training volumes causing an increase in mitochondrial volume, while there is no strong relationship between higher intensity levels (Bishop, Granata, and Eynon, 2014). Since athletes can conduct higher volumes of low-intensity training it stands to reason that a polarized distribution of training would lend itself towards improved mitochondrial volume in endurance athletes.

But, there is much more research to be done in this area.

How to Apply?

I have been applying the principles of polarized training to myself and the athletes I coach for going on ten years now. I definitely don't think it is the only way to train, but I think it can be remarkably effective and can help kick start improvements for athletes who may have been stagnating in their training. For a lot of people it is going to force you to slow down. That's okay, there is no award for running the fastest training miles on a random Tuesday. But, it's a big mindset shift for people that are used to the "no pain, no gain" mentality popularized in sport. Additionally, most people are not going to see immediate improvements; the benefits come after several months. So, it requires patience both during workouts and during the training program. But stick with it and you'll find your low-intensity pace speeding up as well as your race pace. Sometimes you have to trust the process.

One important thing to remember is that polarized training is a principle, not a program. So the application of it will look fairly different depending on the sport, the experience of the athlete, and the time of the season. Training programs always require context! What works well for others might not for you. But based on my experience here are a few good rules of thumb to begin with:

-Perform your longer sessions slower than you think you need to. The gold standards for evaluating intensity are either blood lactate or ventilatory measures attained in a lab, but heart rate and how easy or hard it is to talk are both reasonable proxies (see intensity chart for more). When in doubt, err on going slower.

-Over time allow your low-intensity volume to increase. This should not be a linear increase and there should be variation between weeks (some higher-volume, some lower-volume)

-For every two or three easy session perform one hard workout. The specific parameters of the workout will depend upon your event as well as the part of the season you are in.

-Perform mostly low and high intensity workouts, but mix in some moderate intensity. Especially if you are competing in a sport where the events are of similar intensity to moderate training (marathon running, triathlons etc.)

-When the competitive season is far away focus primarily on low-intensity with a smaller amount of moderate/high.

-As competitive season approaches, make sure that the high-intensity workouts are of a similar intensity to competition.

-All bodies are different, try to learn what works well for *you*

-Training intensity distribution is just one small piece of the puzzle!

Below I will give an outline of the three-zone model of intensity as well as an example training month.

Three Zone Model

Low Intensity: This is a pace that can be held for a long time; for running it will be slower than marathon pace. HR should slowly rise and then stabilize around 75-80% of your max. It's a pace that you can pretty easily hold a conversation at. If you get lactate testing it will likely be a value of around 1.8-2.0mmol. Usually done continuously, or in long intervals with short rest. Examples: 80' continuous with HR under 80% of max.

Moderate Intensity: A pace that is challenging but sustainable. Can be held by most people for a little over an hour. HR rising to 80-90%. Can talk, but not comfortably. Blood lactate values between 2.0mmol and 3.0mmol. Can be done continuously or with long intervals. Example: 40' continuous, 3x18' with 4' of rest.

High Intensity: High intensity workouts have a variety of duration and pace. Almost always split up into intervals to allow for greater total volume. Interval length tends to be 2-10' with duration, total volume, and rest periods altered to fit the goals of the workout. For shorter intervals HR is not a key indicator due to the delay. Pacing will vary depending on the parameters, but try to hold a consistent speed. Talking is very challenging. Lactate above 4.0mmol. Examples: 5x5'/4'rest, 4x10'/3'rest

Here is how I might apply these principles to an intermediate rowing program. This would be somewhere during the winter months building into the pre-season. The athletes have some training experience, but haven't yet built up the ability to tolerate a large volume. In addition to the listed sessions they would be expected to get two weight lifting sessions as well as one cross-training session in per week.

Sample Training Program

I hope this has been illuminating! This is just skimming the surface, but should prepare you for any further exploration you would like to do on the topic of training intensity distribution. If you want to create a training program that makes your athletes faster, helps prevent overreaching, and can be adjusted depending on the time of the season, it is hard to go wrong with polarized training. If you have further questions about the topic please feel free to reach out. Below you can find my references, many of which go further in depth on the topic.

References and Further Reading

Bishop, D. J., Granata, C., & Eynon, N. (2014). Can we optimise the exercise training prescription to maximise improvements in mitochondria function and content?. Biochimica et Biophysica Acta (BBA)-General Subjects, 1840(4), 1266-1275.

Esteve-Lanao, J., Foster, C., Seiler, S., & Lucia, A. (2007). Impact of training intensity distribution on performance in endurance athletes. The Journal of Strength & Conditioning Research, 21(3), 943-949.

Festa, L., Tarperi, C., Skroce, K., La Torre, A., & Schena, F. (2020). Effects of different training intensity distribution in recreational runners. Frontiers in Sports and Active Living, 1, 70.

Laursen, P. B. (2010). Training for intense exercise performance: high-intensity or high-volume training?. Scandinavian journal of medicine & science in sports, 20, 1-10.

Muñoz, I., Seiler, S., Bautista, J., España, J., Larumbe, E., & Esteve-Lanao, J. (2014). Does polarized training improve performance in recreational runners?. International journal of sports physiology and performance, 9(2), 265-272.

Neal, C. M., Hunter, A. M., Brennan, L., O'Sullivan, A., Hamilton, D. L., DeVito, G., & Galloway, S. D. (2013). Six weeks of a polarized training-intensity distribution leads to greater physiological and performance adaptations than a threshold model in trained cyclists. Journal of applied physiology.

Pérez, A., Ramos-Campo, D. J., Freitas, T. T., Rubio-Arias, J. Á., Marín-Cascales, E., & Alcaraz, P. E. (2019). Effect of two different intensity distribution training programmes on aerobic and body composition variables in ultra-endurance runners. European journal of sport science, 19(5), 636-644.

Pla, R., Le Meur, Y., Aubry, A., Toussaint, J. F., & Hellard, P. (2019). Effects of a 6-week period of polarized or threshold training on performance and fatigue in elite swimmers. International journal of sports physiology and performance, 14(2), 183-189.

Seiler, S., Haugen, O., & Kuffel, E. (2007). Autonomic recovery after exercise in trained athletes: intensity and duration effects. Medicine & Science in Sports & Exercise, 39(8), 1366-1373.

Seiler, S., & Tønnessen, E. (2009). Intervals, thresholds, and long slow distance: the role of intensity and duration in endurance training. Sportscience, 13.

Selles-Perez, S., Fernández-Sáez, J., & Cejuela, R. (2019). Polarized and pyramidal training intensity distribution: Relationship with a half-ironman distance triathlon competition. Journal of Sports Science & Medicine, 18(4), 708.

Stöggl, T., & Sperlich, B. (2014). Polarized training has greater impact on key endurance variables than threshold, high intensity, or high volume training. Frontiers in physiology, 5, 33.

Treff, G., Winkert, K., Sareban, M., Steinacker, J. M., Becker, M., & Sperlich, B. (2017). Eleven-week preparation involving polarized intensity distribution is not superior to pyramidal distribution in national elite rowers. Frontiers in physiology, 8, 515