In the lead-up to the Beijing Olympics in 2008 no one predicted the dramatic arrival of Usain Bolt as the world’s fastest man over 100m. He was seen as an excellent 200m runner and, because he was taller than most sprinters, his coach had originally encouraged him to focus on the longer distance of 400m. Then, in the Olympic finals, he ran 100m in a world record-beating 9.69 seconds. One year later, at the world championships in Berlin, he smashed his own record by running the race in 9.58.
Before Bolt, there was a perception that we were nearing the limits of how fast anyone could run 100m. It had taken over ten years for Ben Johnson’s time of 9.79 at the 1988 Olympics – never entered in to the record books after he was found to have been using performance-enhancing drugs – to be beaten. Maurice Greene eventually ran faster in 1999. By 2007, Asafa Powell had gradually whittled the record down to 9.72. Then, out of the blue, came Bolt. A record that had decreased by an average of 74 milliseconds per decade over the previous century decreased in the space of a single year by 140 milliseconds.
Bolt’s world record highlights just how difficult it is to make predictions in sport. Plotting the gradual improvement in the 100m world record over time reveals a remarkably straight line of slow but continual improvement (see chart). If we had followed this line we would have expected Bolt’s Beijing time of 9.69 to happen about now, eight years after he set the record. For Bolt’s Berlin time we’d have to wait until 2030.
What’s more, in 2008 a straight-line fit would have been seen as overly optimistic. Usually we expect to see diminishing returns in record-breaking, with the rate of improvement gradually slowing down. Bolt outperformed not just a statistical model, but an unrealistically positive one. He shattered not only the world record, but also our ability to make reliable predictions about the 100m sprint.
There are two important lessons we can draw from Bolt’s performance. The first is that unexpected changes come from unexpected places. Before Bolt arrived it was assumed that medium height and a strong build, generating explosive acceleration, produced the best 100m runners. Bolt, who was taller and had a more relaxed running style, undermined this conventional wisdom. What he lacks in acceleration he makes up for in the length of his stride. None of the current challengers to Bolt – Justin Gatlin, Jimmy Vicaut and Trayvon Bromell – have a similar physique to him, but we’ll see another sprinter with Bolt’s stature in the future. Perhaps we’ll see a row of Bolt-alikes line up to run the 100m at the Tokyo Olympics in 2020.
The second lesson is more controversial. It is about how we sometimes use numbers to rush to judgments. Even with all the recent drug scandals in athletics, we shouldn’t cry foul just because a result defies previous trends. The progression of the women’s 100m record also contains a Bolt-like event. Before the 1988 Seoul Olympics, at which Ben Johnson was found to be using performance-enhancing drugs, Florence “Flo Jo” Griffith Joyner smashed the women’s world records for both 100m and 200m. Her performances, along with others set during that period, are tainted with accusations of drug use, primarily because of comments made by the sprinter Carl Lewis about changes in her physique, her dramatic improvement during the 1988 season, and her premature death at the age of 38. Drugs were difficult to detect at the time, and many modern-day commentators and athletes have called for the records set during the 1980s and 1990s to be rescinded. But no one has presented hard evidence of Flo Jo using performance-enhancing drugs. Bolt showed that statistical anomalies should not lead us to assume an individual’s guilt.
But they should lead us to expect records from less conventional athletes. In Rio, Jamaica’s Elaine Thompson, who like Bolt has waited until the Olympics to focus on the 100m, is the athlete to look out for. She ran 10.70 in qualifying. In the men’s event, America’s 21-year-old Trayvon Bromell is an exciting prospect. Bromell is shorter and lighter than most of his competitors, which could potentially allow him to lean further forward while simultaneously increasing his step frequency. Having set his personal best this season with 9.84 he is certainly a contender for gold in Rio, and could even push his time towards Bolt’s record.