Background
Polarized training, or “80/20 training” as it is also known, has become a popular training periodization approach in recent years. Initially, observations that elite Norwegian cross-country skiers spend most of their annual training volume at low intensities (i.e., >80% of training volume), very little at modest intensities (i.e., 0-5% of training volume), and then the remainder at very high intensities (i.e., 15-20% of training volume) led to a surge in interest in this style of training intensity distribution among researchers, coaches, and some amateur/recreational athletes (3). This interest has grown into somewhat of a craze among most endurance athletes. Nowadays, many endurance athletes follow this sort of training approach or, have at the very least, heard of it from one of the many voices preaching its superiority as an approach to training intensity distribution for endurance performance success (e.g., Matt Fitzgerald and one of his many “80/20” books).
The rise in popularity of polarized training has brought more attention and research to the topic of training intensity distribution as a whole, which is essentially the way in which an athlete organizes and divides up the time spent at various training intensities across their total training volume. For example, two different athletes that train for 10 hours each week may both complete those 10 hours of training at vastly different training intensities. This ultimately impacts the physiological response(s) that each athlete will experience from their training.
When it comes to training prescription, volume (frequency x volume), type (i.e., running, strength training, rowing, cycling, etc.), and intensity are some of the most important variables to consider as they all impact the specific adaptations that an athlete will experience from their training program. Total training volume can be easily manipulated by increasing or decreasing the frequency of training and the duration of each training session. Training volume may be increased or decreased for many reasons, such as increasing volume gradually over the course of a build into a long-distance endurance event (e.g., a marathon, full-distance triathlon) or as an athlete progresses in their fitness and experience level. Training volume may decrease, on the other hand, during times of recovery, illness, injury, or tapering into an important event.
Training type is usually selected by matching the exercise selection to meet the specific demands of the event that an athlete is preparing for. For example, an athlete aiming to take on a half-marathon should primarily focus on including running sessions in their training plan. An athlete planning to participate in a multi-day Gran Fondo cycling event will take part in primarily cycling sessions. There are of course reasons when introducing “cross-training” activities, or non-specific endurance activities, is important though. If someone is injured and cannot run, they may swim or use a rowing machine to maintain some endurance-specific adaptations while they recover. Another endurance athlete may also include some form of non-specific cross-training in their plan as a way of building fitness without adding in the additional repetitiveness of the activity that they primarily do, such as a cyclist doing some work on a rowing machine occasionally on an easier day or a runner doing some elliptical work on a recovery day to assist in promoting recovery without the impact of running.
Training intensity, which is just as important of a training variable as volume and type, is often one of the most misunderstood training variables. This brings us back to polarized training, which as I mentioned previously, is simply one way of distributing training intensity across one’s overall volume of training. There are other ways of organizing one’s training intensity distribution, of which we will delve into some of these other approaches below. The question I wanted to explore herein, however, is related to the idea that training intensity distribution is static, or something that one “sets and forgets”. In other words, is polarized training always the best approach all of the time?
Is Polarized Training the “Optimal” Training Intensity Distribution for Endurance Athletes?
Personally, I have written on and discussed on podcast episodes the superiority of polarized training as a general approach to training intensity distribution when compared its opposite, training primarily around one’s threshold (also known as “threshold intensity distribution”, “zone 3 training”, etc.). I can summarize the proposed benefits of a polarized training approach as follows:
Performing the majority of one’s training volume at lower, more comfortable intensities allows for the athlete to accumulate a greater overall training volume at a lower injury, illness, and burnout risk when compared to one training mostly at moderate intensities. The accumulation of more training volume consistently over time without interruption due to injury, illness, and burnout is well known to be related to greater success in endurance sport. Therefore, polarized training could be viewed as a superior approach when compared to a “threshold intensity distribution” approach.
This above summary still holds true based on my experience and what some of the current research literature would suggest. For example, a review paper published in 2015 (1) comparing a polarized training intensity distribution to a threshold training intensity distribution among endurance athletes concluded that “Effect sizes for increasing aerobic endurance performance for the polarized training model are consistently superior to that of the threshold training model. Performing a polarized training program may be best accomplished by going easy on long slow distance workouts, avoiding “race pace”, and getting after it during interval workouts.” Someone reading just this paper alone, or in addition to others that paint a similar picture, might conclude that polarized training is the “optimal” approach for endurance athletes.
However, a problem arises if we forget that training intensity, just like training volume and training type, is a fluid training variable that should not necessarily always remain constant or unchanged. Let’s take an athlete training for their first marathon and do a thought experiment. Most people would agree that this athlete should be primarily running when training as this is going to be most specific to developing the fitness this athlete needs to complete a marathon. Most people would also agree that a sensible approach might be to gradually increase training volume from something that is reasonable and manageable initially to something that is increasingly more challenging in order to build, once again, specific fitness capabilities that are going to help this athlete complete a marathon. The specific frequency and duration of each run session is going to be very personal to that athlete but increasing running frequency and/or duration over the course of the training plan is, in general, going to be a good approach. When it comes to training intensity, however, is it best here to just prescribe 80% of their training sessions as easy and the other 15-20% as high intensity (i.e., polarized training) and continue this approach throughout the entirety of the marathon training plan? The answer depends on many factors, and while there isn’t necessarily one right answer, I would argue the wrong answer in most scenarios would be to leave training intensity distribution static and unchanging.
I think a good example of the point I am trying to highlight above comes from a recently published research paper in the journal, Sports Medicine, by Muniz-Pumares and colleagues (2). Researchers in this study took the last 16 weeks of training data from the Strava accounts of nearly 120,000 runners of varying performance levels and reported training characteristics by marathon finish times. The training intensity distribution of runners was quantified using a 3-zone approach, where Zone 1 consists of easy/aerobic intensities below one’s aerobic threshold/lactate turn-point 1 (LT1), Zone 2 consists of moderate intensities between one’s aerobic and anaerobic//lactate threshold/lactate turn-point 2 (LT2), and Zone 3 consists of hard intensities above one’s anaerobic/lactate threshold/LT2 (see Figure 1 for a visual representation of this typical 3-zone model).
Figure 1. 3-Zone Intensity Model*
Importantly, researchers took the training intensity distribution (abbreviated “TID” moving forward) of each athlete and categorized them into one of the following distributions:
Pyramidal Training Intensity Distribution (Pyr-TID)
- ~75-80% Z1; ~15-20% Z2; ~0-5% Z3
Polarized Training Intensity Distribution (Pol-TID)
- ~75-80% Z1; ~0-5% Z2; ~15-20% Z3
Threshold Training Intensity Distribution (Thr-TID)
- <75-80% Z1; >20% Z2; ~0-5% Z3
High-Intensity Training Intensity Distribution (HIT-TID)
- <75-80% Z1; ~0-5% Z2; >20% Z3
These different TIDs are the most common ways of distributing one’s training intensity. These TIDs above are also unique in the placement of the majority of one’s training intensity. As you can see, with Pyr-TID, the majority of one’s training intensity takes place in Z1 and most of the remainder in Z2. With Pol-TID, the majority is again in Z1 with little in Z2 and the remainder in Z3. With Thr-TID, much more volume is accumulated in Z2 with comparatively less volume in Z1 and little in Z3. Finally, with HIT-TID, one spends comparatively more volume in Z3 with little in Z2 and the remainder in Z1. As mentioned previously, the distribution of training intensity plays a large role in determining physiological adaptations that one experiences. So, each TID listed above will yield specific adaptations.
Finally, marathon finish times were broken down into 30-min time ranges. The fastest marathon runners had a finish time of 120-150 minutes. This was followed by runners finishing between 150-180 minutes. The middle group consisted of those finishing between 180-210 minutes. Next were those that finished between 210-240 minutes. The slowest finishers consisted of those finishing in >240 minutes.
Keeping the above in mind, some of the key findings from this paper include:
Training volume across all runners averaged 45.1+/-26.4 km/week.
Training volume was >3x greater (~107 km/week) in runners with the fastest marathon finish times (120-150 min) compared to those with the slowest marathon finish times (>240 min).
There was a strong negative correlation between total training volume in Z1 (i.e., easy/aerobic training intensity) and marathon finish time, with a higher volume of Z1 training volume being associated with faster marathon finish times.
Higher training volume among the fastest runners was achieved by accumulating more volume in Z1
The most prevalent TID approach was Pyr-TID in which the majority of athlete’s volume was in Z1, progressively less volume in Z2, and little to no training volume in Z3.
The proportion of runners adopting a Pyr-TID increased linearly with those accomplishing faster marathon finish times, ultimately reaching ~80% of runners with the fastest marathon finish times (120-150 min).
What can we take away from the above? Well, first of all, it is quite striking to see those runners finishing in the fastest times completed nearly 3x more running volume on a weekly basis than did those with the slowest finish times. This confirms previous research that indicates increases in running volume are, generally, associated with improved running performance. How was greater running volume achieved by the fastest runners? Those with the fastest marathon times completed more running in Z1 in an effort to accumulate more weekly running volume. This also makes sense as, well, in order to run more frequently and at higher volumes, an athlete needs to remain injury/illness/burnout-free to do so effectively. Running in Z1 is far less demanding on the body than is running in Z2 and Z3. Therefore, to accumulate more running volume, it generally makes the most sense to run more at a Z1 intensity level.
Another really interesting finding was that Pyr-TID was the most common way of distributing one’s training intensity in this study, and that this TID approach was increasingly common among faster runners. To me, this makes complete sense based on the fact that these runners were all training for a marathon and the data from this study was from the runners’ last 16 weeks of training leading into race day. Marathon race-pace, for a lot of athletes, is somewhere in Z2. So, many athletes were likely to emphasize lots of Z1 training for their everyday runs and, on their harder sessions, focusing on marathon race-pace efforts. If researchers conducted this same study in, say runners preparing for a 10K or 5K event, many runners may have spent more training volume in Z3 and less in Z2 (something that more closely resembles a Pol-TID) since 10K and 5K race-pace for a lot of runners falls somewhere in Z3!
Conclusions
Altogether, the results from this study demonstrate that polarized training isn’t necessarily always the best approach all of the time. Training intensity, and the way an athlete distributes it over the course of weeks and months, is highly dependent on what the athlete is trying to accomplish. Yes, polarized training is certainly an effective way of distributing one’s training intensity, but for most endurance athletes it should not be the only way in which training intensity is distributed throughout the whole season. Training intensity is a fluid variable, and it should be adjusted depending on the athlete’s goals, race targets, time of year, etc. A good rule of thumb is to indeed complete most of one’s training at a low/comfortable intensity (i.e., 70-80% of volume in Z1). However, what an athlete does with the remainder of their volume (i.e., the other 20-30%) should constantly be evaluated and altered to align with their needs and goals.
References:
1. Hydren, J. R., & Cohen, B. S. (2015). Current scientific evidence for a polarized cardiovascular endurance training model. The Journal of Strength & Conditioning Research, 29(12), 3523-3530.
2. Muniz-Pumares, D., Hunter, B., Meyler, S., Maunder, E., & Smyth, B. (2024). The training intensity distribution of marathon runners across performance levels. Sports Medicine, 1-13.
3. Seiler, K. S., & Kjerland, G. Ø. (2006). Quantifying training intensity distribution in elite endurance athletes: is there evidence for an “optimal” distribution?. Scandinavian journal of medicine & science in sports, 16(1), 49-56.
Happy training and racing!
-Ryan Eckert, MS, CSCS
Do you enjoy our monthly educational content that we create? Not only do we create written content like what you just read, but we have a podcast too where the goal is also to share science-driven, evidence-based information highly relevant to endurance athletes and coaches. We do all of this for free, and we rely on the generous help and support of others to cover some of our basic operating costs for putting out this content. If you would like to help or support, the best way to do so is by becoming a Patreon supporter.