Research in fell running – three papers sampled and summarised

There has not been much serious research into fell running. Whilst writing my latest book I have been digging deep down many avenues for potential sources. One of these has been of academic publications. The following [warning – it is quite a long ramble – a description that has been applied to the Bob Graham Round!] is a cherry pick of oddments from two unpublished MSc submissions and one journal paper. Some of it is good and interesting, and much of it rather shaky in my opinion. I leave you to decide. Full refs are at the end of the blog.
If you know of any other research on the subject DO let me know.

1 ….. physiological needs, characteristics and training methods

MatthewsThe subject of Matthews’ research was Fred Reeves and his daughter Helen. Fred was 36 at the time, and his daughter was aged 12. The method was a series of measurement tests on both of them. The tests included ECG for pulse rate, age, height, weight and sum of skin folds from eight sites. Chest and waist girths, leg length and chest depth were also measured, plus vital capacity, maximum voluntary ventilation, forced expiratory volume, forced expiratory flow, and forced mid-exploratory flow. Fred’s tests were done on a treadmill, Helen’s on a bicycle ergometer. Some of the main data from the above is summarised here:

Fred Helen
Age 36 12
Height 177 150
Weight (kgs) 63.9 37.6
Waist (cms) 77 61
Leg length (cms) 84 71
Body surface area (sq cms) 180 126
Mean skinfold 5.2 8.025
Body fat % 6.64 17
Resting heart rate (bpm) 42 56

My notes are a little sketchy here (I had to view the dissertation in the Leeds Uni library, and at the time had no particular intention to do anything with them, as it didn’t fit with where my writing was going).

Matthews was going down the path of: ‘Heredity, it seems, imposes the limits on the physical fitness that an individual can attain’. Then refers to research on twins, identical and non, sedentary and trained, before quoting Astrand: ‘choose your parents’. He concludes that a father and daughter will have greater genetic variability than non-identical twins (but I noted that in this study there seemed to be no study of the impact of the mother’s genes).

The research went on to consider vo2max. Helen’s vo2max on the lab bike was 42.05, adjusted for treadmill to a ‘suggested’ reading of 50 (bike measures are known to be % lower than treadmill ones). He estimated that her adult potentially could be 70 if undergoing intensive training. Fred’s vo2max was 79. Matthews postulated that Fred could run a sub-4 min mile, as well as a sub 2-15 marathon. Furthermore, he was world class partly because of ‘an unusually high aerobic power/weight ratio’. He then quotes from research in ‘The Physique of the Olympic Athlete.’ By J. M. Tanner. (George Allen & Unwin Ltd., 1964), comparing Reeves against the mean data for a large sample data from different athletic events:

Height cms. Wt kg. Leg length cms.
Fred Reeves 177 63.9 84
Marathoners 171 59.9 81.6
5/10k runners 174.4 60.8 83.2
3k Steeplechasers 179.2 64.8 86.2
800/1500m runners 180.5 68.9 87.7

Next came the results of a survey of 10 leading fell runners (7 had been Fellrunner of Year)

Mean age 31.3
Mean weight 10st 2 men, 8st 12 women
Mean days training 7 days
Age started running <15 4, <21 2, >21 4
Weekly mileage <50 2, 50-70 none, 70 plus 8
More than once a day 8, once a day 2
Train on track 3
Taper down for races 2
Rest/train lightly after race none
Have rest period during year 3
Strength training none
Flexibility work 2
Diet conscious 2
Mean hours of sleep 8
Had a coach 1

Matthews commented that this was ‘a typically British result. Coaching has yet to arrive or be seen as important to the home athlete. It is interesting to note that in international track and field events, involving equipment, all top athletes are coached. In the light of scientific advances in training methods, physical measurements and diet, the need for a coach seems far more relevant.’ [This was in 1981]

Next in my notes come three fairly random comments from the research:
Gregor (in other research) found that ‘an incline of 6% requires 35% more energy than flat. Whereas a similar grade downhill only reduces the effort by 24%.’

Matthews on training: ‘If I was asked to list the single most important attribute a fell runner must possess, I would say simply it is the ability to run downhill. Amazingly, most fell runners use the downhill in the training session to recover. Some, very few I would guess, actually practice downhill running…… So, you should select a very steep scree-laden, boulder-ridden, as near vertical as possible slope and run down it. Go up a little higher – run down that and so on.’

And finally: When discussing rest one respondent replied ‘I rely solely on injuries for rest.’

Acknowledgements on the paper included: ‘I am particularly grateful to Fred and Helen Reeves for their lively participation and to those other fell runners who have given me the benefit of their experiences in replying to the questionnaire.’

2 ….. physiological and biochemical profile

SimpsonThe subjects of this research were from the Clayton and Skipton clubs ‘who gave both their time and blood in unselfish quantities’. The sample consisted of 10 seniors and 12 vets.

They were tested in the lab, in simulated races and in real races (3 of each), including mid-race tests. Pre and post race-weights and heart-rates were taken. The races were Black Lane Ends (1979), Pendleton Hill (1980) and The Three Peaks (1980).

There are a huge amount of references in this paper. For instance ref no 103 is on ‘age and maturity of runners’, while refs 26 and 27 cover ‘vo2max for distance runners’.

Page 65 has information on anthropomorphic measurements of fell runners against marathoners, some of which was from another paper: ‘10,000m and marathon runners are characterised by a low body weight and fat load. Both test groups [of fell runners] were relatively heavy (Vets mean 68kg, Std mean 70.6kg), with skinfolds indicating a mean body fat load of 18.5% for vets and 21.1% for the Std group. This is high in comparison with marathon runners who typically carry less than 12% fat and often show a 7% load. Much of the fell runner’s training is in an environment where low air temperature, wind and rain are the norm. It may be that the protective advantage of a thicker sub-cutaneous adipose tissue is reflected in the skinfold measurements of both groups. Even the most successful national standard subject carried a 19% fat load. This factor may in part account for the relatively low mean vo2max figure for the Std group of 60.5ml/kg/min and that of the Vets of 61.4ml/kg/min. As the subjects are carrying in the order of 1/5 of their body weight as fat, these figures will essentially be lower than those for a similar group of athletes with a lower percentage fat load.’

Ref 105 is quoted as saying ‘10k runners with similar training and similar vo2max show considerable variation in finishing times, variables other than vo2max may contribute to running success.’ Simpson comments that findings of his study seem to agree with this so far as fell runners are concerned.

When it came to field tests he noted that: ‘Ranked mean finishing positions were therefore tested for correlation against ranked aerobic test results. Significant correlations were identified between finishing position, vo2max and absolute vo2 at OBLA (Onset of Blood Lactate Accumulation)’. Furthermore: ‘Where mid-race samples were taken they showed a significant increase on the pre race sample in all cases. However, for this group there were no significant changes in recorded lactate from mid to post race.’ But some courses were uphill/downhill and the midpoint may well have been at the top of the main steep ascent.

Summary of discussion: ‘both groups of fell runners exhibited higher body fat levels than was expected. This may affect their recorded maximum oxygen uptake. The subjects appear to set a race-pace which avoids lactacidosis; running below OBLA except at the high points of hill courses where lactate levels above 4mmol/L are recorded. Lower lactate levels were regained by the end of races. There was a significant correlation between OBLA as indicated by laboratory aerobic tests, and race finishing position. This suggests a positive advantage (in terms of race success) for runners with high vo2max and a high %vo2 at OBLA. However, there must be other factors to race success which are not explained by this work, as the most successful runner does not have the highest vo2max.’

3 ….. analysis of the Bob Graham Round

HayesNormanThe third article is more recent, but for me proved to be less interesting. The stated aim of the research was threefold. The abstract stated that it: tested an elaboration to the well-known Naismith Rule[1] for predicting progress in mountainous country; tested the conjecture that successful athletes in stamina events apply a constant work rate; and estimated the effect such additional factors as darkness or fatigue.

The data analysed related to a sample of 56 successful BGR attempts, 37 clockwise and 19 anti-clockwise. The schedules (ie timings) for these BGR completions were analysed in great detail. This produced some very complicated tables of ‘work rate’ per section of the route, as part of Aim 2 above. I will leave the reader to access the paper to study this detail. As far as Aim 1 was concerned the researchers proposed 5 variations to the Naismith Rule, covering terrain from flat, through gentle downhill to rough, steep downhill (defined as ‘an area of crags requiring rock climbing skills to move about’). Aim 3 was addressed by trying to use the data to give a realistic value to the effect on pace of darkness and fatigue, again by some fairly complicated cross-analysis of the timings on the sectors done in the dark or towards the end of rounds.

In the discussion section the researchers outlined the following conclusions:

  1. Their revised Naismith Rules are a useful indicator of the time taken to traverse a route in mountainous country at a constant work route.
  2. Runners who successfully complete the Bob Graham Round run faster in the earlier stages than in the latter stages. [They justify this ‘obvious’ conclusion by saying ‘it is apparent from many successful runners’ accounts that runners are encouraged by being ahead of schedule: a runner’s state of mind is as important as his/her degree of tiredness in determining work rate’]
  3. The effect of darkness is to reduce a runner’s pace by about 20%, but this average figure hides a considerable variability caused by such factors as terrain and degree of darkness.


1 – Matthews, N. J. (1981), An investigation into the physiological needs, characteristics and training methods of fell running with particular reference to a champion fell runner and his daughter, unpublished Leeds University MSc dissertation

2 – Simpson, D. (1984), A physiological and biochemical profile of the fell runner, unpublished University of Salford MSc thesis

3 – Hayes, M. and Norman, J. M. (1994), Strategy in Fell Running: an Analysis of the Bob Graham Round, Journal of Operational Research Society Vol 45, No 10, pp1123-1130

[1] It should be noted that have been variations to Naismith proposed at various times (eg Tranter’s variations).


Tags: , , , ,

One response to “Research in fell running – three papers sampled and summarised”

  1. Anthony Kay says :

    Steve, I’m halfway through “It’s a Hill,. . .” and I decided to google your name, and found this blog.
    You ask if anyone knows of any other research on the subject of fell-running. Mike Hayes and John Norman have written several other papers, apart from the one you describe. Phil Scarf (Salford University) has also done some. And then there are my two recent papers doing statistical analysis of uphill and downhill pace in various races; there’s no physiology or biomechanics in them (even though I work at Loughborough University, I’m a mathematician, not a sports scientist), but if you would like a copy, let me know at a.kay”at”

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: