The Mind of the Marathon Runner: London 2010

Why was the men’s marathon in London 2010 such a damp squib? The field was one of the strongest assembled for a big city marathon, and the weather, though a little humid, was fairly good for marathon running.  But Tsegaye Kebede was on his own for the last 8 km and the winning margin was 64 seconds.

Not only did the field include the formidable trio of Kenyans, Samuel Wanjiru, gold medalist from Beijing; Abel Kirui current world marathon champion; and Duncan Kibet, the second fastest marathon runner in history with a time of 2:04:27, but also Zersenay Tadese from Ethiopia, who had set a stunning world record in the half marathon in Lisbon five weeks earlier, though he had yet to complete an international marathon; Emmanuel Mutai, second placed between Kirui and Kebede in the world championship in Berlin in 2009; and Jouad Gharib, third in London in 2009.

Despite the fact that Wanjiru had suffered disruptions to his training due to injury in January and had expressed some doubts about the outcome for the race in a pre-race interview, as described in the discussion with Ewen following my posting on 27th April, the strength and depth of the field still appeared to promise a tightly fought finish. This seemed even more likely when a large leading group reached the half way point in a time slower than 63 minutes. 

Then shortly after halfway Mutai increased the pace dramatically, and only Kirui and Kebede were able to stay with him.  By 30 Km, Mutai had dropped back.  Nonetheless, for a few Km it looked as if the ‘end game’ might be a great duel between Kirui and Kebede.  But suddenly Kirui was no longer there.  Unfortunately, on account of the greater interest of the BBC in the women’s race, especially in British hope, Mara Yamauchi, the cameras did not record the moment when Kebede made the decisive break, but for the final 8 Km he had the race to himself, crossing the line in 2:05.19.  Mutai had moved forward again to second place in 2:06.23, and Jouad Gharib was third yet again, in 2:06:55, while Kirui had dropped back to fifth place, and Tadese was seventh, almost 7 minutes behind Kebede.  Wanjiru and Kibet had dropped out.

It is probable that various factors confounded the hopes for a hard fought race to the line.  One possible factor was the disruption of air travel due to the ash from the Icelandic volcano.  This undoubtedly caused a serious problem for Mara Yamauchi, who was forced to make a complex 6 day journey from her home in Japan.  While one can have sympathy with Mara, travel disruption appears a less credible excuse for the elite men, who were airlifted to London by private jet hired by the London marathon organizers at considerable expense.  So what might have been the main factors?

Mind and brain

Two scientific studies add to the growing evidence that the determinants of athletic performance are far more complex that the physiological aspects of physical fitness, such as oxidative capacity of muscles, muscular strength or cardiac output. 

A recent study by Macora and Staiano from Bangor University provided a graphic illustration of the influence of mental evaluation of the task ahead [1].  In the first stage, a group of fit young men cycled on a stationary bike at 90% of VO2max (an average power output of 242 watts) until the point of exhaustion; the point at which they could no longer maintain that power output.  This point was reached typically after about 10 minutes.  The participants were then required to attempt immediately to produce the maximum power output possible in a five second burst.  The peak power achieved during this five second burst was 731 watts, which is more than three times as large as the power output during the sustained test to exhaustion.  This clearly documents the fact that our mind sets limits on the power output our body can achieve, according to the expected duration of the demand for effort.

Even more recently, a study by Gant and colleagues from Auckland University  has demonstrated that sensory stimuli which trigger an expectation of a fresh supply of fuel can act via the brain to increase the power of muscle contraction [2].   Swilling a carbohydrate rich mixture in the mouth without swallowing produced an increase in muscle power not observed after swilling a similar tasting drink without carbohydrate.    Furthermore, the increase in muscle power associated with the carbohydrate-rich drink was accompanied by an increase in brain activity in the part of the cerebral cortex that controls muscle contraction, demonstrating that the effect was mediated by the brain.

What both of these studies show is that signals from the brain set the limit on power output, and these signals can be adjusted by both conscious and non-conscious mechanisms.  The crucial question that these studies raise is : how much can we train the brain to increase power output when it is most needed?  Athletic training largely focuses on developing peripheral physiological capacities such as blood supply to muscles and cytochrome levels in mitochondria.  However, perhaps more attention should be paid to training the central nervous system.

Competitive middle and long distance runners fall into two categories: aggressive front runners and those with a strong finishing kick.   In part, natural physical endowment plays a role in determine which type of runner you might be, but training and mental preparation are also crucial.   Perhaps the most famous aggressive front runner was Steve Prefontaine who cultivated the mental toughness to crush his opponent by setting a murderous pace from the start.  However, perhaps because of my Australian roots, I think I have learned more through pondering three famous athletic duels, each with an Australian link.

Landy and Bannister

My early memory of sport are dominated by the challenge of the 4-minute mile.  Like many young Australians at the time, all my hopes were invested in John Landy.  In the early 1950’s I loved running, though my own involvement was entirely non-competitive.  I simply enjoyed running to and from school.  As the years went by, I looked forward to the day when I would be able to run in the mile on sports day at school, but I never dreamt that I would actually win a schoolboy mile state championship one day.  To be realistic, my lack of dreams on my own behalf was well founded.  I only won that schoolboy championship race because atrocious weather and a water-logged grass track converted the mile into an endurance event, well suited to my preparation based on running to and from school.  While my lack of dreams of glory for myself were well founded, it was almost unthinkable that Landy would not be the first to run a mile in less than 4 minutes. 

It was era when the world’s first commercial jet airliner, the De Havilland Comet was famous for crashing, and the piston-engined Lockheed Constellation with its elegant dolphin shaped fuselage and three finned tail, was on the verge of obsolescence.  International sportsmen often traveled by ship.  As a result we were far less aware of the leading European or American athletes, such as Roger Banister or Wes Santee. Landy filled  the foreground of our awareness.  In the antipodean summer of 1953-54 Landy recorded 4:02.x on four occasions, and then set out for Europe where he would renew his attempts in the European summer, pushed by stronger competition.  When we heard that Oxford medical student, Roger Bannister had broken the 4 minute barrier in May, assisted by his pace-making friends in an event at Iffley Rd, it seemed such an anti-climax.  In June, Landy snatched back the world record with a time of 3:57.9 at Turku in Finalnd.  However the real battle would occur in August 1954 when Landy and Bannister lined up for the mile at the Empire Games in Vancouver.      

As a runner who had done much of his racing far from the more competitive environments of Europe and America, Landy had relatively little experience in tactical running.  In contrast, Bannister had honed his legendary final kick with scientific precision in sessions with his friends Chris Brasher and Chris Chataway   In that miracle mile in Vancouver, Landy decided his best strategy was to establish a good lead by the end of lap three, and was about 10 yards ahead at the bell.  However the crucial moment occurred as the runners came off the final bend and into the home straight.  Landy looked over his left shoulder as Bannister stormed past on his right.  Bannister won in 3:58.8, though both runners recorded a time under 4 minutes. 

Bannister and Landy were both men of outstanding character.  Landy will be long remembered by his compatriots not only for his herculean attempts to break the 4 minute barrier, but also for magnanimously stopping to assist Ron Clark  when he fell during the 1500m final at the Australian National Championships in 1956.  Landy went on to win the race.   Landy achieved a lifetime PB for the mile almost a second faster than Bannister, but Bannister quite deservedly has a much more secure place among the running greats, and he emerged as the victor in that dramatic head-to-head contest, largely because of Landy’s momentary loss of confidence coming into the home straight.  

Pirie and Kuts

In 1956, the international tensions of the deepening Cold War were exacerbated when Russian tanks had rolled into Budapest to crush the Hungarian revolution.  However, the dominant event looming in the minds of many Australians was the Melbourne Olympics.  By this time, Emil Zatopek, triple gold medalist in Helsinki four years previously, was fading and the dominant figures in 5000m and 10000m were Valdimir Kuts from Russia, Sandor Iharos from Hungary and the Englishman, Gordon Pirie.    Kuts, a Red army officer with an aggressive front running style, had captured Zatopek’s 5,000m world record 1954.  In 1955, Sandor Iharos snatched that record from Kuts.  Then in June 1956, in Bergen, Norway, Gordon Pirie achieved a convincing victory over Kuts, while taking possession of the world record with time of 13:56.8.    In 1954 Zatopek’s record for the 10,000m had fallen to Iharos.   The Hungarian held it until Kuts took it in a race in Moscow shortly before the Games with a time of 28:30.4. 

Unfortunately Iharos’ career disintegrated in the turmoil following the Hungarian revolution, and the expectation for Melbourne was a battle between Kuts and Pirie, certainly in the 5,000m and perhaps also in the 10,000m.  Pirie had established his supremacy over Kuts in the 5,000m in Bergen, but Kuts arrived in Melbourne in great condition, fresh from his 10,000m record-breaking run in Moscow.  However Pirie also took pride in his gritty endurance based on prodigious volume of training, much of it at high intensity.  He ran over 200 miles per week.  His coach was the ‘father of interval training’, Woldemar Gerschler.   It is probable that in Pirie’s mind he nurtured the belief that he could crush Kuts in the 10,000m as well as the 5,000m. 

This was not going to be an amateur contest such as we had seen in Vancouver two years before between a privileged Oxford medical student and a gentlemanly Australian.  In the fifties that adjective didn’t seem such an oxymoron – in those days amateur athletics cultivated an Arthurian aura of honour in Australia as in Britain.  But, in the Olympics we expected a ‘bare knuckle’ confrontation between a forthright non-establishment Englishman with an axe to grind, and a Russian army officer who appeared to embody the ruthlessness of a formidable monolithic state. 

The final of the 10,000m occurred first.   Both runners believed that they had the endurance and toughness to prevail in a long surge to the finish, and the pressure built up with Kuts setting the pace. With five laps to go, Pirie took the lead but a lap later, Kuts surged back,  Pirie cracked and Kuts held on to take the gold medal in an Olympic record time of 28:45.6.  Pirie struggled to the finish in eighth place, 64 seconds behind Kuts.  In the 5,000m, the memory of Bergen had been replaced by the raw experience of the battle for mental supremacy a few days earlier, and Kuts again took the gold with Pirie trailing 11 seconds behind him in second place.      

As an incidental footnote to the international tensions of 1956, after the infamous ‘blood in the water’ polo match between Hungary and Russia, it was in Melbourne that the tradition of athletes from different nations mingling with goodwill during the closing ceremony of the Games was introduced, replacing the previous tradition of marching in ranks behind national flags.

Tergat and Gebrselassie

Forty-four years later, the Olympics returned to Australia, this time to Sydney.  I too returned to Australia that year, not on account of the Games but to visit my dying father.  In the intervening years I had too been a medical student at Oxford and had experienced a frisson of excitement running on the Iffley Road cinder track where Bannister had run the first sub-4 minute mile.  But by 2000, I was overweight and unfit.  I was now living in Vancouver.  The old Empire stadium where Bannister had defeated Landy had long since been demolished, but, inspired by the forest trails of the Pacific Spirit Park, I was just beginning to run again.  Shortly after my arrival back in Adelaide, one afternoon I was assisting my elderly mother into a taxi outside the house where I had spent my childhood, on our way to visit my ill father in the Royal Adelaide Hospital, when my elder brother drove past.  He stopped the car about 150 yards down the road, and because it seemed the natural thing to do, I set off unthinkingly at a brisk trot along the road to greet him. As I returned to the cab, the driver grinned at me: ‘With a turn of speed like that, you should be in Sydney, mate.’   Although there were more serious things on my mind that day, that little incident has stuck in my memory. 

The centre of gravity of distance running had moved unequivocally to Africa.   From the midst of an astounding wealth of talent among Kenyans, Ethiopians and Moroccans,  two outstanding figures had emerged to dominate the 10,000m in the closing years of the century, Haile Gebrselassie of Ethiopia and Paul Tergat of Kenya.  Gebrselassie had set a world record of 26:43.53 at Hengalo in 1995, and after losing it, regaining it and losing it again, he had recaptured the record with a time of the on 26:22:75 in 1998, again at Hengalo.  In the intervening time, Tergat had been in possession of the record for almost a year, with a time of 26:27:85 run in Brussels in 1997. 

Although Gebrselassie arrived in Sydney in 2000 as the current world record holder, Tergat was currently in great form, having recorded a time of 27:03:87 in Brussels only a month before Sydney.   He was determined to reverse the finishing order in Atlanta in 1996, when Gebrselassie had narrowly defeated him.   However a review of other previous head-to-head confrontations suggested that Gebrselassie was usually able to call upon some extra reserve when needed.  In both the 1997 and 1999 world championships, Gebrselassie had taken the gold medal and Tergat the silver.

The race was utterly spectacular. A pack of three Kenyans and two Ethiopians loped away from the rest of the field. At the bell, Haile was in a commanding position in second place, with his compatriot Assefa Mezgebu at his side, but all five were running strongly.  Tergat knew of Haile’s capacity to mount a devastating final sprint, whereas most of Tergat’s previous victories over his rival had been cross country events requiring physical strength. It is probable that he was thinking that his best strategy was to launch a strong attack relatively early in the final lap in the hope that his strength would prevail.  It was certainly crucial that he should begin the final sprint before Haile.  250m from the finish, Tergat made his decisive move.  He burst clear of the leading pack by a margin about 1 ½  metres, with Haile in pursuit.  As they sprinted around the curve, Haile slowly narrowed the gap, but on entering the home straight the space between them was still almost a metre and Tergat continued to look strong.  In that heart-stopping final 100m both athletes sprinted with amazing power.  The gap was narrowing at a scarcely perceptible rate and it looked as if Tergat might just hold on to his slender lead.  Less than 10 metres from the line Haile drew level and Tergat showed a faint flicker of weakening.  He dipped towards the tape desperately, but Haile had crossed the line less than a tenth of a second ahead of him, in a time of 27:18.20.


All three vignettes were duels between two fairly evenly matched athletes at the peak of their powers.  In all three instances, the stronger athlete prevailed, but the excess strength was more mental than physical.   The margins were quite different: Tergat crossed the line 0.09 seconds behind Gebrselassie, whereas Pirie was 64 seconds behind Kuts.  While it is probable that Pirie and Kuts were a little less closely matched than Tergat and Gebrselassie, the principle factor accounting for the difference in margins was the fact that the decisive test for Pirie occurred four laps from the end, whereas the decisive test for Tergat occurred only about 4 metres from the line.  According to Wikipedia, Kuts subsequently reported that if Pirie had held on any longer in their titanic battle, he, Kuts, would have submitted [3].

But the key feature of all three vignettes was the crucial point at which the battle was lost.  Landy looked over his shoulder as he came of the final bend with 100 metres to run; Pirie cracked as he and Kuts surged at a ‘suicidal’ pace with 4 laps to go; and Tergat’s posture sagged in a desperate dip only a few metres from the tape.  While the crux can be more easily identified in the body language of the vanquished, the mental state of the victor was equally important.  Bannister had developed his devastating final kick in training with his colleagues, Chataway and Brasher. Kuts had developed the mental strength to sustain a murderous, potentially even suicidal pace, long before the final lap, as he developed his aggressive front-running style.  Gebrselassie had demonstrated repeatedly during the five years leading up to the turn of the millenium that he could always dig a little more deeply whenever required in the final lap of a 10,000m race, and as he followed the powerful figure of Paul Tergat down the home straight in that titanic battle in Sydney, somewhere in the recesses of his brain was the confidence that he had the power to prevail.

Back to London 2010

As the three medal winners from the 2009 world championship in Berlin applied the pressure that demolished the remainder of the elite field shortly after the halfway in London 10 days ago, there were few pointers as to which of the three would eventually be the victor.  Mutai, who had been second in Berlin,  had initiated the surge in London, but after about 8 Km, he apparently sensed that he was not as strong as his two rivals and dropped back, conserving his energy for a final effort that would secure second place here, as in Berlin.  Kirui had been victor in Berlin and had that memory to buttress his confidence.  Unfortunately due to BBC’s preoccupation with the women’s race, I do not know what tipped the balance in favour of Kebede.  Did he have the confidence to surge again, as Kuts had done to crush Pirie in the 10,000m final in Melbourne?  Whatever happened, Kirui, like Pirie fifty-four years previously, cracked, and eventually finished almost three minutes behind Kebede.

The race failed to provide the exciting finish promised by the strength and depth of the field assembled at the start, but in retrospect, the way in which various factors ranging from Wanjiru’s  pre-race loss of confidence due to injury in January; the vagaries of travel though the dust of a volcano cloud; and above all, the way in which the mental strengths of the three medal winners from the world championship in Berlin determined events in the second half, made London 2010 an event to contemplate and savour.  


[1] Macora and Staiano (2010) ‘The limit to exercise tolerance in humans: mind over muscle’ (Eur J Appl Physiol.. [Epub ahead of print])

[2] Gant et al (2010) Brain Research, DOI: 10.1016/j.brainres.2010.04.004


10 Responses to “The Mind of the Marathon Runner: London 2010”

  1. Paul Wilson Says:

    Canute. Great observations as always. I recall there are various studies which indicate that the mind is really adept at dosing our effort to keep us going at what it thinks is a sustainable pace for the distance it thinks we are doing. Now, if we could only trick ourselves … “hey, I am only running 10km this morning (ha, ha, little does my mind know we are running a marathon!!)”

    • canute1 Says:

      Paul, Thanks for your comment. As you are no doubt aware, simply focusing on the distance to the next lamp post is a useful strategy. What I personally find most helpful is focusing on my running form, especially on getting off stance as quickly as possible on each stride, rather than on distance.

  2. Ewen Says:

    Canute, with the Tergat and Gebrselassie race, the mental battle over the last 150m was set up by both runners knowing the other’s strengths. Tergat is thinking that Geb always wins sprint finishes, and Geb has the confidence to win because he’s always won that way.

    It’s a shame the London TV coverage was preoccupied with the women’s race. The Science of Sport blog sheds some light on how Kebede won. After the astonishing 28:53 10k between 20 and 30k, Kebede surged the 21st mile in 4:33 –

    On the study of the stationary cyclists, I’m curious as to whether ‘immediate’ means with zero rest after ‘exhaustion’. Also whether they were sprinting at the end of the exhaustion phase. As sprinting is alactic, one could be exhausted aerobically, then raise an alactic sprint producing high power.

    • canute1 Says:

      Thanks for the information about Kebede’s surge in mile 21. Although a 4:33 mile represents only a small increase in pace compared with the preceding 10Km, any increase in pace at all after 10 Km in 28:53, and with 10Km still to go, must have felt much the same to Kirui as the pressure that Kuts applied with 4 laps to go felt to Pirie in Melbourne in 1956.

      With regard to Tergat and Geb, I agree that the problem for Tergat was that both he and Geb knew that Geb would out-sprint him if they were still in contact with 150m to go. With hind-sight, making his move with 250m to go was not radical enough.
      In view of the fact that Tergat’s training placed quite a lot of emphasis on developing mental toughness, I suspect a better strategy might have been surging at around 8,000m, as Kuts had done against Pirie.

      With regard to the Macora study, you raise an interesting point about a possible switch to alactic metabolism. However, the maximum voluntary cycling power (MCVP) test commenced as soon as the criterion for exhaustion was reached. In the MCVP test, for the initial 2 seconds there is no added resistance to allow the participant to get the revs up to a high level and then resistance is added. The crucial issue is whether or not creatine phosphate levels were already depleted at the point of exhaustion.

  3. Rick Says:

    I think we are only just starting to learn how powerful our minds can be, most of us have a very limiting perception of what is possible!

    • canute1 Says:

      Rick, Thanks, the Radiolab material sounds very interesting; I will set aside some time to listen to it on the weekend

  4. Ewen Says:

    That’s interesting. I wonder if a similar sort of test could be done with runners? Maybe an increasing incline treadmill to exhaustion, then a 100m sprint on a track?

    With Kebede’s 21st mile split, of course, that doesn’t say how that mile was run. Could have been a 2:12 half mile to get away, then easing back to 2:21.

    • canute1 Says:

      Ewen, Thanks for your comment.
      I think probably the equivalent test for a runner should be entirely on the treadmill to ensure that the time interval between reaching the exhaustion criterion and commencing the maximum power test is only a second or two. A possible treadmill protocol might be: set incline and pace to achieve 90% VO2max; take the point at which cadence falls below 50 to be the exhaustion criterion; rapidly lower the incline of the treadmill (with hands on the rail for 2 seconds) while the runner is exhorted to get the cadence back up to 90, and then provide continual exhortation to go as fast as possible for 5 seconds (eg display current pace and offer a reward based on maximum pace achieved pace – possibly within a competition between the participants to make it fun.

      With regard to the London marathon, I agree that the crucial issue is the pace of the surge averaged over about 400m , not averaged over one mile. It is a shame the BBC camera did not record the surge. I have not been able to find any description of the surge in newspaper or blog descriptions. Despite the millions who were watching this event either in London or on television, it appears that few apart from Kebede and Kirui themselves really know what happened. I would be delighted if any readers of this blog could provide any more information.

  5. Ewen Says:

    Yes, that sounds like a good protocol for the treadmill. The only slight problem would be the time the treadmill takes to get up to “sprinting speed.” A second treadmill could be used. Also there’s the possibility of an accident with a fatigued runner attempting to run at top speed.

    • canute1 Says:

      I am not an expert in treadmills, but I assume the maximum rate that a treadmill could get up to speed is set by the rate at which a runner can accelerate A sprinter can get near to peak cadence within about 3 seconds. In the context of the test we are trying to design, I suspect we would have to allow 5 seconds to get to peak. [The fastest treadmill acceleration rate I could find in a quick search is 3 metre/sec/sec, (for a machine custom made by Bonte) . This would require 3.3 sec to reach to 10 m/sec.]

      I agree that it could be dangerous trying to produce peak power output on a treadmill, when exhausted. Maybe it would be possible to design a sling to support for the arms, a bit like the sling that Jack Cody uses to encourage runners to have a tidy arm action, but anchored to a mount above and behind the runners head.
      But this is getting a bit fanciful.

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 )

Google photo

You are commenting using your Google 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 )

Connecting to %s

%d bloggers like this: