The world athletics championships in August saw some stirring performances, but how much more improvement in world records can we expect to see? Progress has been painfully slow in the men's 100 metres, often seen as the blue-riband event of any championship. As long ago as 1930, the world record was 10.3 seconds, and now it is only 9.77 seconds. Reducing the world record by 5 per cent took 75 years. Even cutting it by 2 per cent took 37 years, with the great Jim Hines running 9.95 seconds in 1968. Hines ran under ten seconds on a cinder track, in contrast to the high-tech surfaces used today, so we could postulate that no real improvement has taken place at all since the late 1960s. But using simple extrapolation, the world record will be 9.57 seconds in 2042.
Long-distance running in recent decades has seen a bigger increase in our knowledge of training techniques than the sprints, but even here progress is slow. The current world record for the men's marathon is only 5 per cent less than in 1965. And it is only 2 per cent less than the 1984 record. On this basis, we might expect the first sub-two hour marathon round about 2035.
Perhaps women will improve more quickly? After all, many bog-standard club athletes can beat the women's 1963 world marathon record of 3 hours 37 minutes, and the current record is a massive 38 per cent less than this. On this basis, the current men's record would now be 1 hour 23 minutes.
But it was only towards the end of the 20th century that cultural barriers were overcome and women began serious training in the distance events, so extrapolation is particularly uncertain here. Taken as a whole, the current set of women's world records actually have a longer lifespan than those of the men. There are 24 men's track and field records recognised by the International Olympic Committee, and their median lifespan is 8.5 years. In contrast, the median for the 22 women's events (women are still not deemed capable of the 50km walk or the 3000m steeplechase) is no less than 17 years.
But closer inspection suggests there are some distinctly dodgy records here. Of the 12 women's world records which pre-date 1990, no less than 10 are held by athletes from the former Soviet bloc, some of whom were pumped to the gills with drugs.
One of the most queried women's records is the 29 minutes 31 seconds set for the 10,000m by Chinese athlete Junxia Wang in Beijing. The Chinese have never performed as well outside their own country. But this is 12.1 per cent slower than the men's record for the distance, and the marathon record held by our own Paula Radcliffe is only 8.4 per cent slower than the men's.
In fact, the median difference between men's and women's records in the 17 events which are directly comparable is exactly 12 per cent. So the Chinese record looks perfectly sound.
The smallest difference is the 7.4 per cent in the 100 metres, but the 1988 record of 10.49 seconds held by the late Florence Griffith Joyner is thought by many people to have been obtained by dubious methods.
