2011 was the year that the Kenyans changed the marathon. They made it look like a track race, destroying fields and racing their way to an astonishing collection of victories and records in every Major city marathon. Rotterdam and Boston a week ago didn’t follow the trend, and neither did London, at least in terms of course records, but it was still super fast – Mary Keitany produced the fastest marathon in the world this year, a Kenyan record, and the fifth fastest of all-time (mixed and women only races) with her exceptional 2:18:36. Wilson Kipsang almost broke the course record, missing it by (for him) the now familiar time of four seconds in 2:04:44 (you’ll recall that Kipsang last year missed the world record by the same margin).
However, their performances are noteworthy for the entirely different manner with which they were achieved. And it is Kipsang’s, or rather the men’s race, in particular, that is particularly noteworthy for how it has reminded us of the theme of 2012 – the marathon has struck back. If 2011 was the year when the Kenyans turned marathon running into something of a prolonged track race, then 2012 is the year when the marathon has reminded us all that it’s still a race that requires control, patience and a good deal of respect for how effort is ‘spent’ over its distance. An analysis of the men’s race illustrates this nicely, and comparison with Keitany’s victory further emphasizes this point.
Let’s look at both races, starting, for a change, with the women’s race.
Keitany – patience and building to an extra-ordinary final 15km
Keitany’s record breaking day was built off an extra-ordinary second half after a patient first half. The last time we saw Keitany in a marathon, she was going backwards in New York after starting at World Record pace. She would later describe how she felt better than she really was, and followed that feeling to what was really a suicidal pace, for which she paid but still clung on to third place.
There was no early suicidal pace in London today, but she was responsible for a murderous pace at the end, as Keitany ran a perfect marathon against arguably the strongest field of women ever assembled. Halfway was reached in 70:53, projecting a 2:21:46, which was, at the time, fairly slow given the quality of the field.
It wouldn’t remain slow for long, and the graph below, showing Keitany’s 5km splits (blue line), illustrates just how the pace ramped up from halfway onwards. It wasn’t the same dramatic surge that we saw from Keitany in 2011 (the red line), where she attacked at halfway and reeled of a 32:10 10-km interval (67:52 half marathon pace), but it would end even more spectacularly, as Keitany ran the second half in 67:43! And that was with a progressive acceleration, as the graph below shows.
Unsurprisingly, the rest of the women’s field was being stretched out by Keitany’s pace. A lead group that consisted of 8 was cut progressively after halfway, and ended up being four women with 10km left to run, the Ethiopian challenge dealt with by 30km. That’s when Keitany’s pressure at the front really began to tell, and Florence Kiplagat and Priscah Jeptoo dropped off the back. This left Edna Kiplagat as Keitany’s sole challenger, but Keitany never wavered. She didn’t appear to do anything spectacular – unlike in the men’s race, there was no single kilometer that did damage, but rather a gradual winding up of the pace.
That segment from 35 km to 40 km was covered in 15:45, and is the fastest 5km segment ever recorded by a woman in a major marathon (Paula Radcliffe’s fastest was a 15:47 opening 5km back in 2005). By virtue of that interval, Kiplagat’s challenge was broken, and the world champion fell back by 40 seconds over this interval alone. She would eventually hang on to finish in second, with a sub-2:20 of her own, but Keitany just got faster and faster.
Some other extra-ordinary stats about Keitany’s second half are:
- The final 12.2 km were clocked at 38:43, which is 31:44 pace for a 10km
- The final 7.2 km were run in 22:35, which is 31:22 pace for 10km
- Her final 2.2 km were timed at 6:50, which is 3:06 per kilometer
The circles in the graph above show Keitany’s time gap in 2012 compared to last year’s 2:19:19 victory. You can see that the pace early was slower – 34 seconds lost in the first 5km, then another 8 to 10km and so on. At the 30km mark, Keitany was 42 seconds slower than in her 2011 victory, but then she started to move. The 15:45 “record” interval took her ahead of last year’s time, and then she closed out the final 2.2km in an incredible 6:50, to run 43 seconds faster than in 2011 (not shown on graph).
The 6:50, incidentally, was better by only three other athletes on the day – Wilson Kipsang who won the men’s race, and ran 6:45, and then Martin Lel and Tsegay Kebede who sprinted the finish straight on route to a 6:37! Keitany was simply untouchable in this strongest women’s race ever. What this does is set up the Olympic Marathon beautifully, with Liliya Shobukhova awaiting Keitany. Shobukhova last year won Chicago in 2:18:20, and is the only athlete faster than Keitany in the last 7 years but has been beaten by Keitany (London 2011).
Radcliffe, of course, has run faster than both, but there are serious doubts over her ability to produce anything like what we saw today, and it seems that that is what it will take to beat Keitany, who has now shown herself to have learned the marathon and has the pedigree to win Olympic Gold for Kenya.
As a final word, it’s interesting to note that the spectacular 67:43 second half by Keitany is still slower than Paula Radcliffe’s second half during her 2003 run. In that race, after going through halfway in 68:02 (2:51 faster than 2012), Radcliffe ran the second half in 67:23! Testament to the quality of that performance. But, Keitany’s final 10km is the fastest ever for a woman marathoner, as our friends at Letsrun.com have written.
Men’s race: Wilson Kipsang wins a race of attrition with an incredible mid-race surge
Now let’s look at the men’s race, and keep that graph of Keitany’s in your mind – she started patiently, ran a steady first half and then got faster and faster, culminating in a record 5km interval from 35km to 40km.
Wilson Kipsang did it differently. In the men’s race, it was all about aggression early. Not in the first half, because like in the women’s race, that was fast but evenly paced. They hit halfway in 62:12, after 5km segments of 14:36, 15:00, 14:54 and 14:43. You can contrast this with Emmanuel Mutai’s 2011 course record, where halfway was reached in 62:44. So it was fast.
And then it got faster. Wilson Kipsang, at least from TV pictures, seemed to be the main aggressor, and as the race reached the halfway mark, he blew it wide open. The splits would later reveal why – the 5km segment from 20 to 25km was run in a spectacular 14:09. That’s one of the fastest 5km segments ever measured in a world class marathon (Wanjiru in London 2009, Mosop in Chicago 2011 are faster), and it was responsible for creating massive gaps in an incredible men’s field (Geoffrey Mutai ran a 14:12 segment in Boston in 2011). This meant that London 2012 produced the fastest 5km splits for both men and women, though for Keitany, this came at the end, for Kipsang, the middle…
Kipsang was followed, at first, by Worku and Lilesa of Ethiopia. Abel Kirui was a little slower to respond, but he did bridge the gap, and by 25km, he had replaced Worku in the front three. From 25km to 30km, Kipsang, Lilesa and Kirui continued to work hard on the front, growing the lead over the chasers, who included former champions Tsegay Kebede, Emmanuel Mutai and Martin Lel, to just over 1 minute. At this stage, the front three looked assured of podium places at least, and it was Kirui who did most of the front running between 25km and 30km, looking very strong and full of running.
Then, the next time we saw the race at around 33km, Kipsang was clear, so a split happened somewhere at about 32km. The gap at 35km would grow to 15 seconds, and it was created entirely because Kirui and Lilesa’s pace dropped, and not by a surge by Kipsang. In hindsight, this small gap of 15 seconds at 35 km was the first symptom of an impending implosion for both Kirui and Lilesa.
The graph below shows the 5km splits for Kipsang and Kirui (red and green, respectively) as well as Mutai in 2011. You can see that having been locked together up to 30km, Kipsang held the pace at 14:42 and then 14:43 per 5km, whereas Kirui and Lilesa, still together at this stage, slowed down to 15:00/5km pace. That was enough to put Kipsang clear, and even though he would also slow down, running a 15:11 from 35km to 40km, his lead grew, because Kirui and Lilesa were going backwards by this stage.
The best illustration of Kirui and Lilesa’s difficulties comes from a comparison with Lel and Kebede in the chase. At 30km, the gap was 1:02, and Kirui and Lilesa were locked in battle with Kipsang.
By 35km, Kirui and Lilesa had lost 15 seconds to Kipsang out in front, but their lead over the chasers had grown to 1:42. Then it started to come down, steadily at first, then precipitously.
By 40km, Lilesa was only 18 seconds ahead of Lel and Kebede. He therefore lost 1:24 of his lead within 5km. Kirui was still hanging on, 51 seconds ahead, so he had lost 51 seconds over the same period.
Then over the final 2.2km, things went particularly badly. Lilesa took 8:24 for the final 2.2km and went from third to tenth, whereas Kirui ran the final 2.2km in 8:33 and dropped from 2nd to 6th. After losing 51 seconds over the 5km from 35 to 40km, Kirui then lost an additional 1:56 over the final 2.2km. Partly, this is because he shut it down after being caught, but with prize money at stake, his and Lilesa’s final 7km illustrate how costly the mid-race aggression was.
Kipsang meanwhile was holding the speed a lot better – having run 14:42 and 14:43 between 25 and 35km, he dropped to a 15:11 from 35km to 40km, and then ran just outside 3 min/km to the finish. Having been ahead of Mutai’s record pace from about 16km (see circles in the graph above), Kipsang then drifted outside and just missed the course record, which, it must be said, he probably deserved for the manner with which he beat such a strong field by such a large margin.
In some respects, this London race was similar to what we saw in Boston last week, with those involved in big mid-race surges paying dearly for it at the end. It’s a lesson in “cost-benefit” analysis of marathon running and pacing! The difference between London and Boston, of course, is that in Boston, the eventual winner Wesley Korir was very conservative and did not get involved in surges, whereas London was ultimately won by its main aggressor in Kipsang. However, for Kirui and Lilesa, London 2012 was the same as Boston was for the likes of Levy Matebo and Matthew Kisorio – mid-race aggression has a cost, and the marathon makes sure that cost is paid!
The margins between great and imploding – a fine line
The lesson then, is that the East African strategy of racing the marathon aggressively soon after halfway may exert a heavy price, one that has always existed but that we didn’t notice too much when records were falling left, right and centre in 2011. Today, it was a 14:09 split for 5km that blew the race open, gave Kipsang the victory, but also saw world class runners reduced to speeds of just faster than 4 min/km for the final 5km! In Boston, the heat made this effect even more pronounced, but it’s a lesson in how fine the margins are between extra-ordinary and “mortality”.
Today, in London, Wilson Kipsang lived on that line, and managed to produce ‘extra-ordinary’, winning by 2:07. Mary Keitany ran under that line until the time was right, and then she produced something truly extra-ordinary. But for others, like Kirui and Lilesa, the line was too fine and they crossed it and paid. Having been blown away by the “ease” of the Kenyan dominance of 42.195km in 2011, we are reminded how easily things can go wrong.
For Kirui, then, London was not a good day out. Had he been more cautious, perhaps not followed the 14:09 surge, he might have come through strongly and finished second, far closer to Kipsang. These men are racers, however, and are not interested in steady and conservative efforts to finish second when winning is an option, and this makes marathon running so enthralling. However, Kirui paid for this today.
And the result, his sixth place, means the Kenyan selectors now have to take a “risk” picking him. The same is true for Lilesa. Instead, Wilson Kipsang looks to have secured his spot, and now it’s a judgment call for the other two places. Do the Kenyan selectors go with the beaten Mosop and Kirui? Or do they pick the non-finishing champions of 2011 in Geoffrey Mutai and Patrick Makau? (who bailed at 16km, reportedly with a leg injury). That’s the next big marathon story, and then come the Games, and the clash between Kenya’s best three and Ethiopia’s best three.
2012 – deep high quality year, despite the lack of records and attritional nature
A final word – despite the ‘slow’ Boston performances, and despite everyone but Kipsang being made to look a little more mortal today and over the last few weeks, 2012 is still on course to be the strongest year ever for marathon running in terms of depth of performance. Kipsang’s 2:04:44 today was the seventh sub-2:05 time this year, which equals the 2011 record for most sub-2:05s.
It was also the 14th time under 2:06 in 2012, and that is the highest in history. So while we haven’t seen the course records and clusters of men racing to these times in the Major Marathons, we’re still seeing the continuation of a revolution in the marathon. It’s just not quite as spectacular as it was in 2011.
Among the women, incidentally, the same is true. Keitany and Kiplagat went under 2:20 today, making them numbers 5 and 6 this year, the most ever. Twelve women have broken 2:22 this year, and that’s also a record.
Of course, both the men’s and women’s stats have been helped by that Dubai race in January, where basically half these times were recorded. So the stats are a little skewed, because outside of Dubai, the Major City Marathons haven’t been as spectacular, but history will go on to record 2012 as the deepest year in marathon running, and we’re only in April!
That’s it for the Spring Marathons. Busy time, as always, but hopefully you enjoyed the coverage, both here after the race, and also over on Facebook and Twitter. I don’t usually ask, but since it’s the end of the season, if you’d like to make a donation to support The Science of Sport and our efforts at sports analysis, please do so at the button on the top right of the page.
Until the next big event, which I’m sure is not far away!
Oh, and one final thing – it was great to see Martin Lel sprinting to the finish to claim second place today. Lel, a past champion in London and New York, was the great marathon runner a few years ago, but injury, and perhaps age, have caught up with him, and this was something of a surprise return. He was my favourite marathon runner in around 2007, 2008 and his sprint finish at the end of marathons one of the great sights in running – remember his kick off the final bend to win London in 2008? Today was a moment of nostalgia. Lel isn’t in the Kenyan squad, he wasn’t named to the six man shortlist, but my impartial opinion is that if it was a slow race, I’d have him in it, fitness allowing!