At the 2008 Olympics, the men’s marathon was won by 21-year-old Sammy Wanjiru and the women’s marathon was won by 38-year-old Constantina Tomescu-Dita. That prompted a lot of talk about changes in the age considered “ideal” for fast marathoning, which was usually considered to be late 20s and early 30s. So were these performances outliers, or has there really been a change in elite marathon demographics? Physicist and running blogger Graydon Snider plots some interesting data in an attempt to find out.
The average age was 28.4 before 2000, and 28.3 since – not a big change. A running blog (which I won’t name and shame) claimed that many top runners tend to be around 35. Not true. Graydon also does some more sophisticated analysis like calculating the skew of the distributions, which reveals that older runners may be slightly over-represented in the post-2000 group compared to the pre-2000 group (perhaps because professionalisation means that it’s feasible for runners to keep competing for longer?). And he does the same analysis for women’s sub-2:30 performances, which also haven’t changed much in the two time periods.
So does this mean the much-talked-about phenomenon of younger runners tackling the marathon is just an illusion? Not necessarily. For my recent feature on the prospects for a sub-2:00 marathon (Runner’s World December 2014 issue), I analysed age trends for the top 100 marathoners of each year going back to 1990. When I looked at year-by-year age histograms, it was hard to pick out any discernible pattern. But when I looked at the average age of those top 100, a change jumped out:
It looks to me like Sammy Wanjiru was an outlier when he won the Olympics, but that his example perhaps inspired other younger runners to move to the marathon earlier than they might otherwise have. Prior to 2008, the average age hovered just under 29; since 2008, it’s been just under 28. That’s a pretty significant change in a short period of time. It would be neat to see a similar analysis done for women, to see if Dita’s 2008 win sparked an increase in the number of older women continuing to excel.
One other interesting detail from Graydon’s analysis emerges when he superimposes age-group world records on his histogram data:
What’s interesting is how flat the world-record curve is between ages 18 and 36. Of course, age-group records represent single performances by outstanding individuals. Still, the flatness of the curve suggests that the relative scarcity of elite performances from younger (e.g. 22 and under) and older (e.g. 33 and older) athletes may have as much to do with lower participation rates as with physiological limits. We’ll have to check back in five years and see how those curves have evolved…
Thanks again to Graydon for letting me share his work, and make sure to check out the full analysis.