Sports Med 2007; 37 (4-5): 420-423CONFERENCE PAPER 0112-1642/07/0004-0420/$44.95/0

© 2007 Adis Data Information BV. All rights reserved.

Biomechanical Factors Contributing toMarathon Race SuccessKeith R. Williams

Exercise Biology Programme, University of California, Davis, California, USA

During marathon running, the body’s metabolic resources become depletedAbstractand musculoskeletal stress and fatigue begin to hinder performance, makingefficient running a must. Biomechanical studies of long distance running havetried to identify how body structure and running mechanics interact with perform-ance, economy and injury, and typically have involved groups of subjects. Whilemoderate relationships have been identified, the outcomes include conflictingresults, vague conclusions and unclear consequences. Easily identifiable anduniversally applicable patterns of efficient movement have not been found. Analternative avenue of research is described that concentrates on identifying how anindividual runner’s structure and functional abilities influence performance, econ-omy and injury. It is hoped that when such an approach identifies importantrelationships for individuals, the patterns identified will lead to a more generalunderstanding of the underlying mechanisms.

Marathon-length racing places demands on the cross-sectional studies examining different abilitybody that are different from those for shorter dis- levels of runners or in laboratory studies related totance races. The body’s metabolic resources are metabolic economy. Others have investigated themore likely to become depleted, making efficient relationships between anatomical and biomechani-running a must, and musculoskeletal stress and fa- cal factors and injury susceptibility, on the premisetigue may begin to hinder performance. The long that reducing injuries will lead to better training andtraining runs necessary to stimulate the musculo- performance. While studies have identified moder-skeletal and physiological adaptations needed for ate relationships between biomechanical factors and2–4 hours of marathon running have to be done performance, economy or injury, the overall resultswithout incurring injuries that might limit or halt are disappointing with many conflicting results,training. These demands raise questions concerning vague conclusions and unclear consequences. Therehow biomechanical factors play a role in successful do not appear to be easily identifiable and universal-distance running. Are there optimal movement pat- ly applicable patterns of ‘efficient’ movement thatterns for marathon runners? Are elite athletes suc- will apply to all runners. A more promising avenuecessful because they are able to find their optimal of research may be to concentrate on the individualrunning style while others do not? Are atypical runner in an effort to identify how that person’srunning mechanics poor mechanics? structure and functional abilities influence perform-

Some biomechanical studies of long distance ance, economy and injury susceptibility. We do notrunning have tried to identify aspects of running make all runners wear the same size shoe, for obvi-style associated with better performance, either in ous reasons. Perhaps there is a similar reason why

Biomechanical Factors 421

we should not expect optimal running mechanics to would likely have a much smaller effect, althoughbe the same for everyone. changes as low as 1%, if translated to marathon

racing time, could change times by 1 or 2 minutes.1. Performance A number of studies have provided indications

that most runners are very good at finding move-Analysing running mechanics during race condi-ment patterns that minimise energy expenditure.tions has typically been done to identify consistentCavanagh and Williams[6] found that eight of tendifferences between elite runners and sub-elite orrunners chose SLs that were at, or very near, the SLaverage runners. A complication when examiningthat minimised energy consumption. Morgan et al.[7]

race data is that most biomechanical variables varydemonstrated that it was possible to train runnerswith running speed, making it difficult to comparewho chose an uneconomical SL to run at a SL closerbetween runners where variations may be primarilyto the one predicted to be optimal, with a concomi-a result of speed. Simple measures such as stridetant lowering of V̇O2. This gives some hope thatlength (SL) and stride rate have been studied repeat-when uneconomical running patterns can be identi-edly, but even for those measures, conflicting resultsfied, it is possible to make changes to improvecan be found.[1] For example, while one study foundeconomy, although not without considerable effortthat elite runners had longer SLs at a given speedover an extended period of time.compared with good runners, another found the

Running involves phases where the muscleopposite relationship.[1] Some studies[2,3] have iden-lengthens eccentrically while active during the earlytified a number of biomechanical measures thatpart of support and then shortens during the push-offseem to be associated with elite runners, but rela-phase, leading to toe-off. This cycle of muscle use istionships are usually weak, sometimes contradicto-termed the stretch-shortening cycle (SSC). There arery, and not easily applied to any given individual. Inwell-documented benefits to SSC showing that itaddition to comparing between different levels ofhelps to reduce metabolic energy costs comparedrunners, there have also been a variety of effortswith what pure concentric muscle action would en-aimed at identifying differences based on gender[3]

tail. SSC is thought to be an important contributor toand race.[4]

running economy, and while SSC has been studiedextensively, it is still not known how to identify2. Economywhether a given runner uses this movement pattern

An alternative research mode used to evaluate effectively, whether it is a neuromuscular patternbiomechanical traits that might influence perform- that can effectively be trained or modified,[8] and ifance has involved running economy, submaximal so, how best to do that.oxygen consumption (V̇O2) while running at a givenspeed. The assumption is that if a runner could lower 3. InjuryV̇O2 by some mechanism, such as by ‘improving’running mechanics, performance times should low- Lower extremity overuse injuries are a constanter, since a faster pace could be maintained at a given worry and frequent problem for marathon runners.metabolic effort level. An interesting study by With proper training, the body shows an amazingEgbuonu et al.[5] had subjects run either holding ability to build stronger tissues capable of with-their arms tightly behind their back or while trying standing increasingly greater repetitions and force.to maximise vertical oscillation, with the intent of As with performance, some studies in the literatureshowing how submaximal V̇O2 increased with rath- show relationships between specific aspects of run-er extreme changes in mechanics. The average in- ning style and injury, or between anatomical factorscrease of 1.6 mL/kg/min (≈4%) provides a possible and injury, while others show no difference.[1] Forupper limit for how much economy might be altered example, rearfoot pronation following footstrike isby a change in mechanics. More subtle changes one of the most studied parts of the running motion

© 2007 Adis Data Information BV. All rights reserved. Sports Med 2007; 37 (4-5)

422 Williams

and it is frequently cited as a cause of knee and foot difficult for scientists, physicians, coaches and train-injuries. Nigg et al.[9] found greater pronation in ers to identify causal relationships.runners with tibial tendonitis, supporting such arelationship between pronation and injury, but 4. Evaluating the Individualothers have found no relationship.[10]

Figure 1 shows an example of the confounding While many studies have concentrated on analy-relationships that can be found when trying to relate ses of groups of runners, it may be more meaningfulpronation to injury. Groups are shown for subjects to consider distance runners as individuals. Rela-with a history of shin splint problems, Achilles tionships between biomechanical factors and injury,tendonitis or who had never been injured. Conven- economy or performance might best be understoodtional wisdom would assume greater pronation by examining various parameters on a case-studywould be associated with a greater incidence of basis. By carefully examining biomechanical as-injury, but that relationship doesn’t hold for individ- pects of running style for an individual in relation touals among these runners. Figure 2 shows a similar measures of anatomical structure (e.g. range of mo-inconsistency for subjects running in two different tion at various joints, arch height), functional abili-shoes. While six of the eight runners showed the ties (e.g. flexibility, muscular strength), injury histo-same or greater pronation in shoe 2 compared with ry, shoe wear patterns, training methods and othershoe 1, two subjects markedly reduced pronation in factors, it can be possible to tie various parametersshoe 2. For whatever reasons, some interaction be- together to build a more cohesive view of howtween anatomy and running mechanics caused the various components are related.shoes to influence pronation in different ways for Consider the example of an elite female runnerdifferent runners. who had a history of bilateral Achilles tendon

There do seem to be relationships between problems, persistent bilateral plantar fasciitis and abiomechanical measures, anatomy and injury, but recent third metatarsal stress fracture. Biomechani-the relationships can be complex and do not always cal measures showed that compared with other eliteapply to different runners in the same way. What females she had an unusually long SL, an abnormal-might cause an injury in one runner can have no ly high amount of knee flexion during the supportinfluence on another, making it that much more phase and very limited pronation during footstrike in

−20

Mean = −8.7°

Shin splintsAchilles tendonitisNo injuries

−15

−10

−5

01

Max

imum

rea

rfoo

t pro

natio

n (°

)

47 Elite female runners

6 11 16 21 26 31 36 41 46

Fig. 1. Maximal rearfoot pronation for 47 elite female runners showing groups with chronic Achilles tendonitis, shin splints or with a history ofno injuries.

© 2007 Adis Data Information BV. All rights reserved. Sports Med 2007; 37 (4-5)

Biomechanical Factors 423

provide insight into the interconnecting factors thataffect performance, economy and injury. If enoughinformation becomes available showing relation-ships for individuals, trends may become evidentthat can lead to a more general understanding of theunderlying mechanisms.

Acknowledgements

Keith R. Williams is a consultant to Kilpatrick Stockton,LLP Atlanta and a contracted researcher for Puma.

References1. Williams KR. The dynamics of running. In: Zatsiorsky V,

editor. Biomechanics in sport: performance enhancement andinjury prevention. The encyclopaedia of sports medicine. Vol.IX. Oxford: Blackwell Science, 2000: 161-83

2. Williams KR, Cavanagh PR. Relationship between distance

0

−2

−4

−6

−8

−10

−12

−14

−16

−18

−20Shoe 1 Shoe 2

Max

imum

rea

rfoo

t pro

natio

n (°

)

Fig. 2. Maximal rearfoot pronation for eight runners in two differentshoes. Two runners show the opposite trends with footwear com-pared with the other runners. running mechanics, running economy, and performance. J

Appl Physiol 1987; 63: 1236-453. Williams KR, Cavanagh PR, Ziff JL. Biomechanical studies ofboth feet. Force platform data showed very low peak elite female distance runners. Int J Sports Med 1987; 8: 107-18

ground reaction forces for a forefoot striker. Anthro- 4. Enomoto Y, Ae M. A biomechanical comparison of Kenyan andJapanese elite long distance runner’s techniques [abstract no.pometric measures showed her to have a relatively852]. XXth Congress of the International Society of Bi-

rigid foot, probably a cause of the limited pronation omechanics; 2005 Jul 31–Aug 5; Cleveland (OH)5. Egbuonu ME, Cavanagh PR, Miller TA. Degradation of runningat footstrike and a potential link to her plantar fasci-

economy through changes in running mechanics. Med Sciitis problems. Since subtalar pronation can help with Sports Exerc 1990; 22: S176. Cavanagh PR, Williams KR. The effect of stride length varia-shock absorption, her minimal pronation may have

tion on oxygen uptake during distance running. Med Sci Sportsbeen related to why she flexed her knees so muchExerc 1982; 14: 30-5

during support, as the increased knee flexion would 7. Morgan D, Martin P, Craib M, et al. Effect of step lengthoptimization on the aerobic demand of running. J Appl Physiolalso help to reduce maximal forces. The increased1994; 77: 245-51

knee flexion may have also affected ankle move- 8. Spurrs RW, Murphy AJ, Watsford ML. The effect of plyometrictraining on distance running performance. Eur J Appl Physiolments and stress on the calf muscles, and combined2003; 89 (1): 1-7with her forefoot strike pattern may have contribut-

9. Nigg BM, Luethi S, Stacoff A, et al. Biomechanical effects ofed to the Achilles tendonitis problems. Although pain and sportshoe corrections. Aust J Sci Med Sport 1984; 16:

10-6stress fractures are often associated with higher than10. Messier SP, Davis SE, Curl WW, et al. Etiologic factors associ-normal loading, her extreme forefoot position at ated with patellofemoral pain in runners. Med Sci Sports Exerc

1991; 23: 1008-15footstrike coupled with minimal pronation may haveput enough stress on the metatarsals to cause the

Correspondence: Keith R. Williams, Exercise Biology Pro-stress fracture despite low ground reaction forces.gramme, University of California, 196 Briggs Hall, Davis,

Tying biomechanical and other information to- CA 95616, USA.gether in ways similar to that just described can E-mail: krwilliams@ucdavis.edu

© 2007 Adis Data Information BV. All rights reserved. Sports Med 2007; 37 (4-5)