Journal of Applied Physiology

The two-hour marathon: who and when?

M. J. Joyner, J. R. Ruiz, A. Lucia


In this Viewpoint we ask if information about the physiology, genetics, and empirical history of elite endurance performance can provide insight into the question of “who” will break the 2-h marathon barrier and when this might happen. We also identify several physiological questions that we believe need attention.

The current world record in the men's marathon is 2:03:59 (Gebrselassie 2008). This record has fallen by more than 16 min since the early 1950s after high volume/year round training was adopted widely. Except for the 1970s, the record has fallen by ∼1–5 min/decade since 1960 when Africans entered international competition. Improvements since 1980 likely also reflect increased prize money and competitive opportunities that allowed top athletes to earn a living running. Figure 1 shows the history of marathon times and projected improvements. Using times from 1960, the open squares suggest it will take 12–13 yr to break 2 h assuming an ∼20-s reduction per year. If times from 1980 are used, the filled squares suggest it will take 25 yr assuming an ∼10-s reduction per year. Consistent with the idea that marked improvement is likely, empirical models of running times suggest that the men's world records for the 10,000 m and half marathon are equivalent to a marathon time of ∼2:02–2:03 (5, 21).

Fig. 1.

Progression of world record times in the marathon since the late 1920s. The rapid fall in record time in the 50s and 60s likely reflects: 1) the widespread adoption of high volume/year round training after WWII and 2) the participation of East-African runners in international competition starting in the 1960s. There was limited progress during the 1970s, but the record has fallen more than 5 min over the last ∼30 yr. On average, there has been ∼20 s reduction per year since 1960. □ indicate if this rate of improvement continues, a time under 2 h could occur in 12–13 yr (by 2021–2022). ■, show that if only data from 1980 are used, a time under 2 h would occur in ∼25 yr based on an estimated improvement of ∼10 s per year. The recent increase in the number of high profile races on fast courses that offer substantial prize money may also contribute to faster world records in the near future.

Physiology of the 2-h Marathon

The physiological determinants of distance running performance (V̇o2max, lactate threshold, and running economy) have been used to develop a model of marathon performance (9, 10). Elite marathon runners typically have V̇o2max values ranging from ∼70 to ∼85 ml·kg−1·min−1. These individuals can sustain running speeds that require 85–90% V̇o2max for more than 1 h, and these factors along with knowledge of the oxygen cost to run a given speed (running economy) provide a reasonable estimate of marathon pace (9, 10). When outstanding values for these three key variables are used in this model, a sub-2-h marathon seems physiologically possible.

While there are many possible combinations that might lead to elite performances, it appears that extremely high values for V̇o2max and outstanding running economy are rarely seen in the same person (9, 10). East African runners do not have particularly exceptional values for V̇o2max or lactate threshold, but generally have outstanding running economy (13, 14, 23). The classic study of Pollock (19) showed that elite distance runners who focused on the marathon had lower V̇o2max values and better running economy that those who focused on shorter races. On the basis of these data and other anecdotal reports, it appears that whoever breaks 2 h for the marathon will have exceptional running economy (2, 4).

In this context, there is clearly a need for more information about the relationship between V̇o2max and running economy and the physiological explanation for the relationship if it exists. There is evidence that V̇o2max and gross mechanical efficiency are inversely related in cyclists and influenced by muscle fiber type (16). By contrast, running economy seems more related to mechanical factors, including vertical displacement and so-called braking on foot strike (11, 24). Exceptional running economy might also provide two important physiological advantages. First, fuel use would be lower and perhaps glycogen depletion delayed. Second, metabolic heat production would also be lower, potentially reducing thermal stress. To our knowledge these potential advantages have not be studied extensively.

What Will the 2-h Marathoner Look Like?

Forty-one of the 50 fastest marathons have been run by Kenyans or Ethiopians (1). Importantly, the mean height and weight of the 30 runners (29 Africans) who have broken 27 min for 10,000 m is 170 ± 6 cm and 56 ± 5 kg, with only one runner greater than 178 cm or 70 kg (12). Additionally, most of these athletes had exposure to high altitude and significant physical activity early in life. In this context, small body size has a favorable effect on V̇o2max; however, less is known about its influence on running economy (7).

From these observations other questions emerge: 1) does exposure to the combination of high altitude and physical activity early in life lead to pulmonary adaptations that reduce the incidence of arterial desaturation seen during heavy exercise in elite athletes (3, 5, 15, 16) and 2) would the reduction in metabolic heat production along with a favorable body weight-to-surface area ratio have the net affect of reducing thermoregulatory stress during periods of prolonged, intense exercise? While these questions might be difficult to study, small differences could be decisive when races are won and records set by very small margins. However, there are examples of “big” runners like Paula Radcliffe, Ron Clarke, and Derek Clayton who have been highly successful. Importantly, Radcliffe and Clayton are known to have superb running economy, and Radcliffe's running economy improved dramatically over time, providing at least some evidence that this factor is “traininable” (8, 19).

Genotype: Probabilistic Versus Deterministic

Genetic factors may limit or enhance the possibility of running a very fast marathon. At present much of what is known comes from association studies, with the angiotensin converting enzyme (ACE) I/D and α-actinin-3 (ACTN3) R577X gene polymorphisms having been studied extensively. The ACE I allele is theoretically associated with improved cardiovascular function during exercise, and could also favor muscle efficiency (26). While there is an overrepresentation of the I allele in the best Spanish marathon runners (sub 2:09 marathon performance) (15), the ACE I/D polymorphism is not associated with the success of the best elite endurance runners worldwide, including Kenyans (25). The association between the ACTN3 R577X variation and elite “power” athlete status is strongly documented (27), yet this is not the case for endurance running (28).

Beyond potential genotype/phenotype associations (which are yet to be clearly established in elite marathoners), the task of quantifying the genetic contribution to elite marathon performance is challenging. A record holders's phenotype results from the combined influence of hundreds of genes, epigenetic factors, and non-hereditary environmental influences. Using algorithms that take into account the combined influence of several candidate gene variants associated with endurance performance [i.e., the so-called “total genotype score” (TGS), ranging from 0 to 100], it appears that genetic factors increase the possibility of becoming a marathon champion (22). For example, a Caucasian individual with a TGS value above 75 has ∼5 times greater chance of achieving elite endurance runner status compared to those with a TGS below 75. Yet, less than half of the best Spanish marathoners have TGS values above 75; and, using this approach it is estimated there are nearly 6 million Spanish individuals with the “genetic” potential for elite marathon performance. Whether having the best possible TGS (i.e., 100) increases the odds of breaking 2 h is unknown.


Whoever breaks 2 h will likely have outstanding running economy and small body size along with exposure to high altitude and significant physical activity early in life. However, neither of these factors nor any specific suite of genotypes appear to be obligatory for a time this fast. Current trends suggest that an East African will be the first to break 2 h. However periods of regional dominance in distance running are not unique to the East Africans: athletes from Finland, Eastern Europe, Australia, and New Zealand have all had extended periods of success at a range of distances (17). From a physiological perspective, more information is clearly needed on the relationship between V̇o2max and running economy and the influence of running economy and body size on thermoregulation and fuel use.


No conflicts of interest, financial or otherwise, are declared by the authors.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.
  22. 22.
  23. 23.
  24. 24.
  25. 25.
  26. 26.
  27. 27.
  28. 28.