How to Train (Part 4) – Threshold Training

How to Train (Part 4) – Threshold Training

In this penultimate post on how to train – I will be discussing the key aspects of threshold (THR) training, after having covered both; polarized training (PT), high-volume training (HVT), and high-intensity interval training (HIIT) in previous posts. As with the previous three models, there is plenty of evidence that suggests THR training is a brilliant way to train for all types of endurance athletes (which I will cover now), however the trick is to know when to apply which type of training model in your training – which I will discuss in my final post on how to train.


THR is a training model that makes use of a combination of Long, Slow Distance running, and Pace/Tempo training (see How to Train Part 1). Most researchers agree that in order for a training model to be classified as “THR”, one needs to spend approximately 50 – 60% of their training time doing Long, Slow Distance running and the remainder of their training time should be spent doing Pace/Tempo running (and small amounts of Speed/Interval running) (Hydren & Cohen, 2015) (Stoggl & Sperlich, 2015). It is worth noting however, that a universal definition of THR training is not really a thing, illustrated by the fact that the THR model is defined by Kenneally et al. (2017) as, “ which higher volumes (>20%) of running are performed in zone 2 (see table 1 below) than other models”. The apparent benefits of doing THR type training is adding a large proponent of training time at “race pace” which simulates the muscle fiber recruitment patterns of a normal competition performance, as well as improving running economy and improving one’s lactate threshold (Haff & Triplett, 2016).


THR is defined by Stoggl and Sperlich (2015) as, “Training performed mainly at an exercise intensity corresponding to the lactate threshold or second ventilatory threshold…often defined as zone-2 training”. Now for those of you following this blog, this corresponds to approximately your zone 3 / zone 4 pace in your Joe Friel FT test. The zone-2 referred to by Stoggl and Sperlich is according to the three zones posited by Hydren and Cohen (2015) as adapted below:


Table 1: Training Intensities of Three Zones adapted from Hydren & Cohen (2015)

Varying Definitions of Intensity Zone 1 Zone 2 Zone 3
Rating of Perceived Exertion (RPE) / 10 ≤4 5 – 6 ≥7
RPE / 20 ≤13 14 – 16 ≥17
Ventilatory Threshold (VT) ≤VT 1 VT 1 – VT 2 ≥VT 2
Blood Lactate Concentration (mM) <1 2 – 4 >4
Joe Friel’s Zones 1 – 2 3 – 4 ≥5


Recently, it appears that many elite athletes are shifting away from THR training towards a more PT model, with three recent studies documenting this shift in various high level endurance athletes/teams (Olympic speed-skaters, competitive cross country skiers, and international rowers) (Hydren & Cohen, 2015). The shift in Olympic speed-skaters was further highlighted by Stoggl & Sperlich (2015) who described the Dutch Olympic speed skating team as having shifted away from a more THR type model leading up to the 1970 Olympics (40/40/20% Z1/Z2/Z3) to a more PT model in the 2010 Olympics (80/12/8%). I feel like now is a good time to mention that in several intervention studies comparing these four different models, THR training has been found to induce positive change in several key endurance variables for endurance athletes including; 40km time trial performance, peak power output, power at lactate threshold, high-intensity exercise capacity, 10km performance, time to exhaustion, and peak performance (Stoggl & Sperlich, 2015). The only argument being made by the researchers above is that other training models (mostly the PT model), has greater positive effects on these endurance variables than a THR model.


There are varying reasons for this shift away from a THR model, including; athlete’s enjoying a PT model more than a THR model, a lower potential of over training in a PT model, and the sheer stress of a high proportion of training time spent at high intensity (Hydren and Cohen’s zone 2-3) on the athlete’s body which negatively affects recovery (Hydren & Cohen, 2015). Furthermore, an excessive amount of training time spent at moderate to high intensities can have a negative effect on one’s autonomic nervous system (the bodily systems that you have no conscious control over like breathing, your heart beating, and your digestion), stress on the endocrine system (Kenneally et al. 2017) and a greater demand on carbohydrate fueling (Stoggl & Sperlich, 2015).


Now you might ask why a greater demand on one’s carbohydrate fuelling system would be a negative thing – simple physiology! Low intensity exercise (the majority of exercise done in a high volume training model or polarised training model), is fuelled largely by fat, and humans have so much fat stored that they can (theoretically) last on this fat for extremely long periods of time. A large proportion of carbohydrates are used to produce energy when you do moderate to high intensity exercise (as opposed to fat), which is because although we don’t have much carbohydrate stored in the body, carbohydrate is much quicker to provide us with the energy we need to keep up with the demands of the higher intensity exercise than fats are. This is actually a really amazing mechanism, and while it’s great to be able to produce energy quickly, it doesn’t last very long. Therefore, we physically aren’t able to exercise for long periods of time at high intensity – even if we eat right during exercise. Now (and this is important), when elite athletes are required to put in extremely high mileage, or volume of training in order to get a positive training effect, it is reliant on them being able to fuel that type of training. As such, this apparent “negative” of THR, may be of greater importance to elite runners clocking big weeks of training (think 120 – 200km per week), rather than their novice or recreational counterparts who may see a benefit from a greater proportion of higher intensity training. This is another great example of knowing when, and for who, to use which type of training model and even Stoggl and Sperlich (2015) state that THR may be a training model more applicable for untrained and recreational runners.


Now let me throw out a curve ball…and it is a big one. In the study by Keneally et al., (2017), the researchers state (similarly to the previously mentioned researchers) that, “When THR training has been compared (to PT, HIIT or pyramidal training models), it consistently proves to be less effective in the studies available”. Alright so it’s conclusive – three of the most comprehensive research studies ever done on training intensity distribution (training models) all agree that THR training is not effective or is no longer a thing…guess who uses THR training? Kenyans! And not only Kenyans…the world’s best. They use THR training, and earlier it was suggested that THR training is only for untrained or recreational runners – what is going on here!? The coach of these runners (including Florence Kiplagat, Moses Mosop, Abel Kirui, and Wilson Kipsang) is Renato Canova (a big sport science advocate by the way). Canova has coached at the very top, for a very long time – and a lot of his insight is available to the public, just Google his name! Canova makes use of high volumes of THR training (as opposed to most other training models which have a very low proportion of Pace/Tempo workouts) which he describes as “specific race pace” (Keneally et al. 2017). As previously stated, in order to get a performance increase in elite athletes you really have to work hard. I mean Eliud Kipchoge almost broke the 2hr marathon barrier for the first time in the existence of mankind last year! If you want to get a performance increase in him (25 sec would get him to that 2hr mark by the way), you have to work damn hard, or do something miraculous. This is very different to your average recreational runner. Let’s assume the human limit in the marathon is X, right now Kipchoge is pretty close to X, he’s at X – 1 let’s say. Your average recreational runner is at X – 1 000 000 – so it’s much easier to get a performance improvement out of Joe Soap as it is to get one out of Kipchoge. And yet…again…the best marathon population in the world uses THR training. Let me try and make sense of this.


The flaw with all of the studies prior to this is that they did not take the period of training into account. In other words, researchers typically take around 30 runners, split them into  a PT and THR group, and analyse the effects of the training model on key endurance performance indicators (like running economy, VO2MAX or lactate threshold). The logic is inherently flawed. Does this mean that because 8 weeks of PT in these 30 runners induced a greater response, that they should use PT for the next 52, 104, 156, or even 208 weeks? No, it simply means that for that period of time, they saw a greater improvement in a particular variable when using one type of model as opposed to another. This improvement does not look at the bigger picture, the phase of the season, the goals of the athlete and coach, the training they had done prior to the study, or what they will be doing immediately after it. We know for a fact that change is almost always good – and so using one model for 1, 2, 3,or 4 years will start yielding diminishing returns! What we can learn from coach Gonzalo is that the key is more in the “how and when” of training models than which model is best. Coach Gonzalo typically makes use of a block of PT, and then uses a progressively greater THR model when approaching competition season (I.e. a marathon, a series of events, or a world competition) (Keneally et al. 2017). The concept that Gonzalo is using here is that because marathon pace lies in zone 2 (THR), his runners must spend more time training in this zone when building towards an event and less time at zone 1 (low intensity) or zone 3 (high intensity).


In other words, different training models may prove more effective at different phases of your training season. It is the stitching together of all of these models that is truly important, not which model is the best. There therefore is no optimal training model, and if there is – it has not been established yet (Keneally et al. 2017). In my final post I will attempt to explain several ways you can stitch all of these training models together to make a coherent, sensible training plan for yourself. Your goal should be to make the best training program available to you, using all of the tools you have available to yourself – rather than sticking with one model and assuming you will continue to perform better week by week. Thanks for reading – looking forward to the next post? Let me know in the comments below.




Haff, G. G., and Triplett, N. T. (2016). Essentials of strength training and conditioning. Human Kinetics. 4(20) p. 559-581.


Hydren, J. R., and Cohen, B. S. (2015). Current scientific evidence for a polarized cardiovascular endurance training model. Journal of Strength and Conditioning Research. 29(12) p. 3523-3530.


Kenneally, M., Casado, A., and Santos-Concejero, J. (2017). The effect of periodisation and training intensity distribution on middle- and long-distance running performance: a systematic review. International Journal of Sports Physiology and Performance.


Stoggl, T. L., and Sperlich, B. (2015). The training intensity distribution among well-trained and elite endurance athletes. Frontiers in Physiology. 6(295) p. 1-14.

1 Comment
  • […] training and designing training programs for my clients. Let’s start with what I touched on in Part 4 – I don’t believe there is any one best type of training or training model. My belief is […]

Post a Comment