Irish Mountain
Running Association

Distance running training Kenyans 

Distance running training –

How élite Kenyan runners point the way to peak 10k performance


The dominance of distance running by East Africans is a well-established phenomenon. The current IAAF world ranking points system in the 5,000 and 10,000m categories features no fewer than 17 Ethiopian and Kenyan runners in the top 20 ranked athletes. There has been much speculation by coaches and pundits about the reasons for this superiority, which has been variously attributed to genetics, sociological factors and altitudinous environment.


The last issue of Peak Performance reported on a British study examining the relative contributions of nature and nurture to the success of Ethiopian athletes (PP190 December 2003, p10). And now French researchers have added to the growing literature on East Africans with the first well-conducted study of the physiology and training patterns of Kenyan athletes (1).


The team, led by famous running physiologist Veronique Billat, analysed the training diaries and physiological profiles of 13 male and six female Kenyan 10k runners. The major significance of this study is that the subjects were élite athletes competing on the international circuit. A problem with many previous physiological studies has been that the athletes tested were sub-élite, not training as hard or performing as highly as world-class athletes, making it difficult to draw accurate conclusions about what contributes to élite performance. What Billat’s study manages to do is highlight some small but significant differences between élite athletes with distinct training habits.


The study focused on training time spent at three discrete physiological paces:


·                     vLT (velocity at lactate threshold). Billat defines vLT as the pace at which the blood lactate concentration rises by 1 mmol/L to between 3.5 and 5mmol/L. This is higher than at a steady running pace, when lactate levels would be stable between 2 and 3 mmol/L. Nevertheless, vLT is maintainable, since lactate levels do not escalate rapidly, causing fatigue. Physiologists and running coaches have identified vLT as the pace conferring optimal aerobic fitness benefit. Training at this pace leads to an increase in vLT, enabling people to run faster without boosting blood lactate levels. Training at vLT is also known as tempo running, and normal run duration is between 30 and 70 minutes;


·                     vVO2max (velocity at maximal oxygen uptake). This is a very intense pace which can be maintained for only about six minutes, with blood lactate levels around 8-10 mmol/L. Previous research by Billat has shown that vVO2max is associated with superior 10k performance and that training at this pace improves both VO2max and the economy required to maintain pace at this intensity. Training at vVO2max takes the form of interval workouts – e.g. 20 x 400m or 6 x 800m;


·                     vΔ50, the intermediate pace between vLT and vVO2max, which is, as this study confirms, very close to 10k race pace. Training at vΔ50 involves long repetition interval sessions – e.g. 4 x 2,000m - with short recoveries.

The Kenyan athletes in this study can be split into three groups, two male and one female, each with slightly different training habits in terms of kilometres run at these three different paces, as shown in Table 1, below.


Table 1: Analysis of group training diaries





Total weekly distance (k)




vLT weekly distance (k)

10.9 (6.9%)

25.4 (14.4%)


vΔ50 weekly distance (k)

6.8 (4.3%)

2.4 (1.4%)

10 (7.9%)

vVO2max (or above) weekly distance (k)

7.8 (5.0%)


4.8 (3.8%)


All other weekly distance not specified in the table was run at less than vLT, e.g. 90-minute runs at an easy pace.


M1 was one of two male groups, whose members performed faster-paced interval sessions, with a significant proportion of weekly kilometres run at vΔ50 and vVO2max pace. By contrast, members of the second male group (M2) focused on training at vLT, with no interval sessions at vVO2max and a greater overall weekly training distance. The F (female) group completed significantly shorter weekly distances than the men, but included fast-paced intervals sessions at vΔ50 and vVO2max pace.


The physiological profiles of the athletes in each group are summarised in Table 2 below. These profiles were determined by means of a step test performed on a running track, at ambient temperatures of 19-22°C and wind speed less than 2m/s. The athletes started running at a pace below vLT (F at 14 kph and M1 and M2 at 16 kph) and continued at this pace for three minutes. After 30 seconds’ rest, when blood lactate samples were taken, the athletes completed another three-minute run 1kph faster, and so on until exhaustion.


Table 2: Physiological profiles and 10k pbs





Weight (kg)




10 km PB (mins)




10k velocity (kph)




VO2max (ml/kg/min)




vVO2max (kph)




vΔ50 (kph)




VLT (kph)





Oxygen consumption was measured by means of a portable gas analysis system fitted to each athlete, and vVO2max was taken as the pace at which oxygen consumption did not go up.


It is evident from the table that there were differences between the two male groups in terms of physiological profile and 10k personal best, although it is not possible to say for certain that these differences were due to variations in training. The differences between the male and female groups were assumed to be largely gender related.


M1 athletes were significantly faster than M2 athletes over 10k, a feature which is associated with superior VO2max, vVO2max and vΔ50. Statistical analysis showed that the superior vVO2max of the M1 athletes was related to their greater weekly distance at vVO2max and also that vΔ50 and vVO2max were the two biggest predictors of 10k performance.


This important research confirms Billat’s previous findings of the importance of training at vVO2max and the association of this intense pace with élite 10k performance. It would seem that for 10k racing vVO2max has a higher currency than vLT, probably because 10k pace is at greater than vLT pace – i.e. at vΔ50.


Obviously, pushing up your vVO2max will automatically increase your vΔ50, this being an intermediate point between vLT and vVO2max. Also, by focusing on vΔ50 and vVO2max you learn to produce the force required to boost economy at race pace, and speed.


Winning in modern international distance racing seems to require a burst of speed at the end of the race, just when you would expect to be slowing down. According to the venerable South African running physiologist Professor Tim Noakes, aerobic training alone cannot prepare distance runners for this feat; instead, training needs to focus on developing the neuromuscular system to produce the running speed required.


Evidence to support Noakes’ theory was found in the current French study. A number of athletes in the M2 group were unable to reach a true VO2max during the step test, as they fatigued too quickly at the fastest running pace. This was probably due to their lack of high speed training and a consequent failure to produce the muscular force required to run the highest speeds in the test. Clearly, their performance was not limited by oxygen consumption, as this had not yet reached plateau. Thus it was not their cardiovascular systems which failed them.


The lesson here is that to run at your target race pace – and faster – you have to train at these paces rather than simply training your aerobic system. The M2 group ran a greater weekly distance and focussed nearly all of their higher intensity training at vLT, yet they performed worse over 10k than the M1 group, whose aerobic training was relatively neglected.


Nevertheless, the M1 group’s vLT did not suffer from this lack of attention, as it was just as high as that of the M2 group. If you glance back at Table 1 you will notice that M1 and M2 completed a similar number of high intensity weekly kilometres – around 27k per week. The key difference was that M1 split the high intensity distance fairly evenly between vLT, vΔ50 and vVO2max, while M2 performed almost all of their high intensity training at vLT.


It does seem to make physiological sense to train at all three of these paces. vLT training will optimise your aerobic system, vΔ50 training will help you to be economical at 10k pace, and training at vVO2max and faster will help you to generate the force required to speed up at the end of races and develop a greater VO2max capacity.


Interestingly, the Kenyan women in the study trained in a similar fashion to the M1 group but did no training at all at vLT. It is possible that their vΔ50, and therefore their 10k pace, could be boosted by adding in sessions at vLT.


The evidence of this study is that élite Kenyan athletes have very impressive physiological profiles, but previous research suggests that their Caucasian counterparts are capable of developing in a similar way.


In a commentary accompanying the published research, Professor Noakes argues that the ‘X factor’ of performance is the ability of the brain to produce the force required to run at a given pace required, with the cardiovascular system assuming a secondary importance. The fact that the athletes from the slow training group fatigued at high running speeds before reaching their maximum oxygen uptake during the step test would seem to support this theory.


As that great breaker of barriers Sir Roger Bannister wrote in 1956: ‘Though physiology may indicate respiratory and cardiovascular limits to muscular effort, psychological and other factors beyond the ken of physiology set the razor’s edge of defeat or victory and determine how closely the athlete approaches the absolute limits of performance.’



1.                   Medicine and Science in Sport and Exercise, 2003, 35(2), 297-304 and 305.