High Performance Interval Training

High Performance Interval Training
Dr. Gordon Wright 

ABCC Senior Coach




During the 2001 season Stuart Dangerfield broke Graham O’Bree’s 11-year-old 10 miles time trial Competition Record when he turned in an 18 mins 19 secs ride on a course in Yorkshire: the fastest time trial ever ridden in the UK under RTTC competition rules. One week later he won his third National 25 miles Time Trial Championship by more than two and half minutes. Later in the season he won his third Isle of Man International time-trial, did the second fastest ever 10 miles TT in 18 mins 25 secs, and took his fourth British Cycling Time Trial Championship. He then went on to win the National 10 miles Time Trial Championship. In his build-up to the competition record ’10′ and his National ’25′ win in particular. Stuart made extensive use of a form of interval training we have been developing and refining for some three years.

Pyramid Interval Training


The Levels
of Training

These levels were devised by Peter Keen, head of the British Cycling World Class Performance Program, from extensive Kingcycle test data.

Level 1 
More than 45 bpm below measured max heart rate.

Level 2 
35-45 bpm below measured max heart rate.

Level 3 
15-25 bpm below measured max heart rate.

Level 4 
Less than 15 bpm below measured max heart rate.

These levels have since been restructured into seven training zones, as shown in the Heart Rate Training Zone Calculator

The method, which I have called Pyramid Interval Training, was based on a programme of ‘layered intervals’ where different interval durations and intensities were used in each layer with very long recovery periods to ensure interval efforts were maximal. This type of training approach is very hard if undertaken with full commitment, but it appears to be very effective for a rider already at a high level of specific fitness to find that bit extra for the special occasion. In Stuart’s case his UK summer racing and training was also underpinned by a large volume of Level 2 quality endurance training in Australia in the previous winter (see box on the right for an explanation of the Levels of Training used). Our view is this large base of aerobic work enabled him to undertake a substantial volume of this particular form of interval training without any serious downsides. His capacity to absorb and recover from intense interval work was high, with devastating effect in shorter distance time trials.

Specific Muscle Fibre Actions
One of the particular features of training for any sport is that the training effect is very specific to the sport and particular muscular action. This has been known by exercise physiologists for many years, but the full extent of the performance implications seems to have been slow in getting through to many racing cyclists and some coaches. What it means is that how you train muscles determines what they can do well. If you run it does not help cycling directly – we all know that and most people would put it down to using different muscles. But that is only partly true – you use almost all the muscles in your legs to run as well as cycle. The difference is the various muscle fibres are recruited in a different pattern or sequence, and building muscular fitness by running has very little specific carry over to cycling. When the British runner Sebastian Coe set the world record for 800 metres it is unlikely that he would have been able to ride a ten mile time trial at any more than a modest club time trial pace. But given some six months to adapt to a cycling action then the story would have been very different.

To understand why such differences exist we need to consider muscle actions down at the level of individual muscle fibres. Muscles are built up from ‘bundles’ of muscle fibres which must be bio-electrically activated to produce movement. In the large muscles of the thigh for instance there will be many thousands of muscle fibres all potentially capable of being activated to produce muscular force which can be transmitted into pedalling action, ie cadence or angular velocity. But it is only through the repeated and specific action of cycling itself that encourages progressively more and more muscle fibres to be recruited into a pedalling action.

Two key processes are taking place. The physiological development of muscle fibres that enables them to produce more force, and a neurological conditioning and signalling process that sequences more of them into the specific cycling action at any one moment in time. Overall it is a process that takes a long time to develop and become efficient – indeed years to reach a very high level of output. Switch to running and the well developed fibres will not be able to function at all well in a coordinated manner for a running action and fibre recruitment will be low until a new adaptation pattern is established. So one of the keys to successful physiological conditioning for any particular sport is to develop and recruit as many fibres as possible for the very specific and particular muscular actions required by training in very sport specific ways.

The Road to Higher Performance


Muscle fibre types

Fast twitch – type Ilb
These fibres have a high force generation, but are quick to fatigue.

Fast twitch – type Ila
These fibres also have a high force generation, but have some resistance to fatigue.

Slow twitch
These fibres have a lower force generation but are highly resistant to fatigue.

As already mentioned one of the objectives of training is to selectively recruit and activate the highest proportion possible of those motor units that can become directly involved in a sustainable pedalling action. As example, the large frontal thigh muscles are the major power houses of cycling action. Each of these muscle groups is composed of all three types of muscle fibre, but in varying proportion from one individual to another. Endurance training promotes development and activation of the slow twitch muscle fibres and endurance performance improves, as the aerobic capacity increases. Intense interval training will activate and develop both types of fast twitch muscle fibres and high power output and anaerobic performance improves. Although fast twitch fibres fatigue comparatively quickly, with repeated doses of high quality intense training they develop a significantly improved ability for lactic acid buffering [see the reference by Adele Weston].

Acidosis is a localised process affecting individual muscle fibres and motor units. So it is reasonable to postulate that even during intense effort a proportion of fatigued fast twitch fibres will recover quickly. They will then be able to come back into the action and contribute to the high energy processes again whilst other fibres fatigue then rest and recover. This process would be intermittent, but with repeated high intensity and specific conditioning it would be coordinated with the action of the mainly slow twitch fibres. Empirical evidence and experience with Stuart Dangerfield strongly indicates this process and conditioning can be achieved by a large volume of very intense work done in the very specific cycling action and sport discipline that is required for particular competitions – in his case short distance time trials.

The Pyramid Interval Program


A full Pyramid
Interval set

Warm up for at least 15 minutes.

Phase 1
Between 8 and 10 flat out sprints of up to 15sec on a gear large enough for full power development but small enough to keep up good cadence [120+ rpm]. There should be a long recovery period (at least three minutes) between each interval.

Ride easy for 10 minutes for full recovery before Phase two.

Phase 2
Between 6 and 10 one minute very intense intervals are undertaken at level 3 to 4. Recovery period between each one should be 5 to 6 minutes of easy riding. Gearing should be as in phase one – a gear large enough to keep the power on for the full minute, but small enough to keep up a good cadence for the whole one minute effort [110+ rpm].

Ride easy for 10 to 15 minutes for full recovery before Phase three.

Phase 3
Three to four 2.5 mile endurance intervals are undertaken at 1 0-mile TT pace at least – higher if possible. The gearing should be that used in a 10 mile TT. Fast leg speed throughout each interval is important [100+ rpm]. The recovery periods should be at least 6 minutes to allow for full recovery. Each interval must be ridden with full commitment and very high motivation. Distance is strongly preferred to time, as it is more motivational. Each one of these endurance intervals should be treated as a mini-time-trial.

Ride easy for 15 to 20 minutes on the way home for a shower.

This full session will take around two and half hours. Riders must drink a good volume of weak carbohydrate fluid throughout the whole session. They should also drink a considerable volume of carbohydrate fluid [5% strength] after the session to aid faster recovery. A full session should only be undertaken when the rider is completely fresh. The day after should be a recovery day of one hour easy riding.

A half set of pyramid intervals can be undertaken the first time the method is used to allow the rider to adjust to this intense from of training. So 4 sprints, 4 one minute efforts and 2 endurance intervals. The recovery gaps should be the same as for a full set.

The system of interval training that appears to produce the required activation of the key muscle fibres and improved muscle buffering capacity is through what I call ‘Pyramid Intervals’. The pyramid principle has been developed essentially from the notion of ‘layered intervals’ which I first came across in the early 1970′s and is attributed to early ideas by Malcolm Firth. The principle in layered intervals is that you start with long endurance intervals, work up in three layers, and finish with the very short sprint-type intervals. In each layer the ratio of interval work to recovery follows the fairly standard and classical patterns of short recovery periods. The method that Stuart Dangerfield and I have developed turns this on its head, for reasons I come to a little later. Basically it works as follows.

After a really good warm up period of at least 15 minutes of riding and progressively raising the pedal cadence the session starts with intense intervals.

Phase 1
Repeated all out sprints up to 15 seconds in duration are followed by a long recovery period of at least 3 minutes of slow, riding for active recovery. These sprints are repeated a number of times followed by around 10 minutes of easy recovery riding. These short maximal intensity sprints recruit the fast twitch type b (quick fatiguing) muscle fibres which are the ones which require the highest training intensity for activation. Depending on the level of fitness and motivation between 6 and 10 such interval sprints are recommended. Follow this with some 10 minutes of easy riding to allow for full recovery.

Phase 2
One minute all-out efforts (around low level 4) are followed by at least 4 to 5 minute rest periods between each interval while riding slowly to recover. Between 6 and 8 such intervals are undertaken in this phase. These slightly less intense intervals activate the fast twitch type a (fatigue resistant) muscle fibres, and also activate a high proportion of slow twitch muscle fibres by causing a maximal aerobic overload on the cycling muscles. At this stage all three muscle fibre types in the cycling muscles are in state of activation. Follow this with an active recovery period of around 15 minutes easy riding.

Phase 3.
Two and half mile, or five minute, intervals are carried out with at least a six-minute active recovery period between each interval. Again the effort needs to be very high, but sustainable for the duration – at the very least level three but preferable much higher. That is approx 75% to 80% of the maximum power output that can be sustained for 1 minute, or between 90% to 95% of maximum heart rate. Any less and the desired result will not be achieved. Three to four of these long intervals are carried out during which time all the fibre types are now encouraged to work together in a highly coordinated manner, even though the general level of fatigue will be high. The session is completed by a warm-down ride home.

During such sessions Stuart does not monitor his heart rate. It is all done on feel and each interval is ridden at the maximal effort he can sustain for the duration of that interval effort. And one of the reasons we use very long recovery periods is because they are needed to ensure each subsequent interval can again be ridden at maximal effort. We analyse heart rates later to see the quality of the work done. Whilst we seek very high levels of acidosis in the interval effort we aim for very good clearance during the recovery periods. My view is the high ‘average acidosis’ levels found in more traditional interval methods can lead to too much muscle damage, as there is insufficient recovery in the rest interval and acidosis remains high for too long. During sprint intervals Stuart will be close to his maximum muscle power of 950 to 1,000 watts, In the one minute power intervals [uphill in his case] he will hit around 650 to 700 watts. In the fast endurance intervals he will generate some 500 watts for just under 5 minutes. So maintaining around 460 watts during the competition record ’10′ was perhaps not such a huge task after many repeated doses of our Pyramid intervals.

Training Options
You can vary the number of intervals done at each phase in a Pyramid session according to your level of fitness, but the principle should hold, that is the most intense efforts first, the longest duration last. The very long recovery periods are in my mind crucial to the whole process to allow very good recovery so each interval effort can be of the highest quality whilst still maintaining rhytmical good form an high cadence. For time trial preparation I strongly advocate the endurance intervals in phase 3 are done on a flat and fast piece of road to encourage the rider to go as fast as possible with a high cadence and good rhythm in the tribar position. Always with the wind direction and never against the wind.

The idea is to go fast – very, very fast – and learn ‘real speed’, not fight against the conditions. For Stuart Dangerfield all out speed, power and a fast pedalling rhythm in these sessions is all important. Throughout such training sessions I advocate frequent fluid intake to keep well hydrated and the session should be followed immediately by a good fluid intake with about 6% to 8% solution of glucose polymer to help promote fast recovery. For elite riders who complete a full pyramid session there is also a case to be made for lacing the post training drink with a small amount of branched chain amino acids to aid earlier muscle protein resynthesis.

My view is that three beneficial physiological effects come from this form of training. First, the improved motor unit coordination at high intensity in the specific cycling action results in more fast twitch fibres being recruited to sustain higher power levels. Secondly, the increased buffering capacity of muscle fibre acidity at high power output levels enables fast twitch fibres to be used for a longer duration before fatigue, but also to be able to switch back into action after recovery. These effects almost certainly assist each other. Thirdly, using high cadence levels at high force achieves higher levels of sustainable power. This last effect leads to considerable ‘motor learning’ by rehearsing the time trial action at very high levels of power output. Good evidence that the fast twitch quick fatiguing fibres are very active during the long endurance intervals comes from the observation that Stuart can feel his thigh muscles begin to visibly swell during the effort. This is akin to the ‘pumping iron’ effect much sort after by body builders when they ‘pump up’ the muscles by many fast repetitions. This pumping up action is caused by a large volume of lactate accumulation in the muscle which causes fast twitch muscle fibre swelling. Fortunately this phenomena quickly disappears when the effort is ceased. The swelling Stuart observes is a clear measure of just how hard he is able to push himself during the longer intervals and the significant contribution made by his fast twitch fibres.

Pyramid Interval Training in Practice
To be most effective this type of interval work should be used only when a rider is already in a very good overall state of both aerobic and anaerobic fitness. If not then they will find it very difficult to complete the sessions at the level of intensity and volume required to achieve the desired result. For an average to good club rider I would only recommend one pyramid style session each week [maybe two shorter ones] for a specific number of weeks [normally 4 to 6] during the racing season for specific targets. Pyramid sessions should only be attempted when a rider is fully rested. So the day after a rest day or easy riding day is most ideal. If such a fierce session is undertaken in a fatigued state from other training, or racing, then the benefits will be seriously reduced because it is just not possible to reach the levels of intensity required.

Unfit riders just will not get any real benefit and they run the risk of seriously over-training. It has been described by Stuart as the interval method from hell but he knows the benefits he gains from this approach. It can be viewed as ‘shock training tactics’ for the already very fit rider. There is no doubt this is a very hard way to train, but it is very effective in the quest for higher performance cycling fitness especially for the very experienced racing cyclist looking for that something extra to give a competitive edge. If you are very motivated to succeed then this approach can produce results which can be quite dramatic. But it must be monitored very carefully. I would be very wary about prescribing it for young riders under 18 years of age who are still developing physically. They really should not engage in such arduous training methods.

The programme in the box on the right of this article gives just one suggested session of pyramid intervals. The actual number of intervals at each level can be varied according to the level of ability of the rider and their stage of development. Initially quite small numbers might be used until confidence and ability increases, then the numbers can be progressed. Also as with any interval training method the rest periods can be varied and might be even longer in the early stages. The skill of the coach in prescribing very arduous training methods of the type described here is to know your rider well and know how much intense training he or she can absorb in a given time and how quickly they can recover before they can repeat the dose. The methods described here have been used extensively by Stuart Dangerfield to support principally his high performance time trialing. But there is no reason at all why the specialist road rider should not also achieve significant performance benefits by using Pyramid intervals as part of their training programme. In my opinion too many road riders rely far too much on purely aerobic endurance ability.

I would not recommend such methods to older veterans [say 50+] without them first having a complete medical check including an ECG and their physicians full approval to undertake such hard and intensive training. But vets vary enormously as we know and some can ride at a very high levels of performance. Those who have been in regular racing for some long time [months at least] should find these methods beneficial, but the general message should be apply with caution and a detailed knowledge of the individual.

Concluding Thoughts
If it were possible to patent a training method then I would register the process and system I have described here. You can’t do this so with Stuart’s approval we are happy to share it with the world at large and Stuart is the proof of the benefits that can accrue. The benefits in Stuart’s case also come from the effort he is prepared to put into his interval sessions. His motivation to train hard is very high and he will push himself to extreme efforts: far harder than your average club time-triallist. In the six week period leading up to the 2001 National 25-mile Time Trial Championship he undertook some 15 structured sessions of pyramid intervals in various combinations [numbers of intervals in each phase etc.] depending on the day of the week and where he was in his training cycle. He is now a strong and very durable athlete and he is able to absorb a high volume of this intense work. If you choose to use this approach don’t expect to become a competition record breaker, but with careful planning and control and very good recovery days you might see some dramatic results for already very fit riders who are very committed and motivated. But be careful: the risk of serious short-term overtraining is high with this method. As a final thought I can say the experience of working with Stuart Dangerfield has taught me what many of us have suspected back down through the years: if you want first class athletic results you have to train extraordinarily hard. The trick is to do it smartly.

Some Research Support
There is good research evidence available which supports the use of a number of the principles embodied in the type of intervals advocated in this article. Several research articles show the benefits of endurance intervals of around 5 minutes duration. Westgarth-Taylor [1997] and others showed that from just six weeks of twice weekly interval sessions comprising from six to eight intervals of 5 minutes duration at 80% of maximal power with one minute recovery a group of already endurance trained cyclists were able to improve maximal aerobic power by almost 5%.

In a similar study with the same protocol Adele Weston [19971 showed that a three-times weekly interval programme with 5 minutes intervals over a six week period at 80% of maximal power improved muscle buffering capacity by around 14% on average. This enabled 40 km time trial times to be significantly improved.

Nigel Steptoe and others [1998] investigated the effects of different interval training protocols on cycling time trial performance. The protocol which produced the greatest power gain and improvement in 40 km time trial times was three weeks of six session of intervals training with each session comprising 8 by 4 minutes endurance intervals at 85% of maximal power.

These three studies alone give strong support for an intensive interval programme comprising 4 to 5 minutes duration [or 2.5 miles] at an intensity of at least 80% of maximal power to significantly improve 40 km TT times. Note in all three examples the effort in endurance interval work was way above level 3. In Stuart’s case the endurance interavals are ridden at or close to 100% of max ramp test power output.

A study by Bertil Sjodin [19921 showed the neuromuscular benefit of intense interval sessions with very long recovery periods relative to the work period. The argument is that if the aim is to improve speed and power, i.e. neuromuscular efficiency, then the recovery period should be long enough to ensure full recovery before the next work interval. This ensures each work period can be of the highest quality. A further benefit with long recovery periods is to achieve very high lactate levels in the next work interval which the body is then able to clear efficiently and utilise in other tissues.

A study by Tim Noakes demonstrated that a group of black South African runners had superior fatigue resistance in long distance events because of the superiority of their type Ila muscle fibres [fast twitch fatigue resistance]. This was due to their strong preference for undertaking a substantial volume of their training at greater than 80% of V02max. The argument is that this results in type Ila fibres that have developed a more enhanced aerobic capacity and produce less lactate. In other words they gradually become stronger versions of slow twitch fibres.

Noakes etc al. ‘Fatigue Resistance of Elite Black South African Runners’, Journal of Applied Physiology, Vol 75(4), 1822 – 1827, 1993

Nigel Steptoe et. al. ‘Effects of different types of interval training programmes on cycling time trial performance’, Med. Sci. Sports & Excerc. Aug 1998, pp 737 – 741. 1999

Adele Weston et. al. ‘Skeletal Muscle buffering capacity and endurance performance after high intensity interval training by well trained cyclists’ Eur. Journal Applied Physiology, vol 75 pp 7 – 13. 1997

Bertil Sjodin, ‘Maximal Intensity Intermittent Exercise, Effect of Recovery duration’, Int. Journal of Sports Med. vol 13(7) pp 5218- 533, 1992.

Westgarth-Taylor el. al. ‘Metabolic and performance adaptations to interval training in endurance trained cyclists’, Eur. Journal Applied Physiology, vol 75 pp 298 – 304. 1998.


Copyright © Association of British Cycling Coaches 2012