Justin Wolfers’s excellent post on Usain Bolt‘s extraordinary 200-meter race mentions in passing that “it is only a fairly recent phenomenon that the 200-meter typically yields a faster average speed than the 100-meter sprint.” We’re living in a topsy-turvy world where the world-record pace is faster on a longer distance than a shorter distance.
When Bolt set a new world record in the 100-meter race last May, lots of people declared him to be the “World’s Fastest Man.” But to my mind, he didn’t legitimately take that title — neither then nor last Saturday when he lowered his own 100-meter record to 9.69 seconds at the Olympics.
Even after Bolt’s Olympic 100-meter, Michael Johnson had a strong claim to be the world’s fastest.
One of the amazing things about Johnson’s 1996 world record in the 200 meters is that his pace or average speed for the 200-meter record was faster than the average speed for the fastest 100 meters. Johnson, covering twice the distance and running the curve, was able to sustain an average speed of 9.66 seconds per 100 meters. At the time, Donovan Bailey‘s world record in the 100 meters was 9.84 seconds.
So to my mind, Michael Johnson became the world’s fast human in 1996. His act made it clear that the difficulty of having to accelerate from zero in the first 100 meters could outweigh the disadvantages of additional distance and the curve. (This cool graphic shows that Michael Johnson wasn’t the first 200-meter record holder to have a faster pace than the 100-meter record; Tommie Smith pulled off a similar feat in the 1960′s.)
The world record in the 100-meter race has been lowered or matched eight times since 1996 (including times that have been later voided for doping), but no one in the 100-meter — including Usain Bolt with his current 9.69 record — has ever run at a lower average speed than Johnson’s 1996 record.
Bolt’s 200-meter record earlier this week is important because he can now truly claim to be the world’s fasted human — averaging 9.65 second per 100 meters.
Now that Johnson and Bolt have overturned the simple notion that average speed declines with distance, it becomes slightly harder to figure out what distance race in track would maximize average speed. Would it be 10 meters, 50 meters, 150 meters, or 300 meters?
It’s also far from clear whether Bolt’s new average speed record will stand. Bolt ran 9.69 seconds in the 100-meter race without giving his full effort for the last 20 meters. He might be able to run 100 meters at an average speed that is faster than his own 200-meter average speed. Bolt may not have many other close rivals, but the lens of average speed gives him a new way to compete against himself.
and 
. You can also watch it on 
According to some “theoretical” splits developed for Michael Johnson’s race (http://myweb.lmu.edu/jmureika/track/splits/mj200.html) he reached is highest speed between 60 and 70 meters at 11.6m/s, although his average speed continued to improve until 190m, so at least according to one model, that is your answer for the distance at which to show highest average speed. For highest instantaneous velocity, 60 to 70 meters seems to be the widely accepted distance. For comparison, Bolt achieved a top speed of 12.2m/s during his 9.69 run.
Also at the 2008 olympics, the Jamaican 4×100 relay team (of which Usain Bolt is a member) set the world record to 37.10s, meaning that at least one of the four runners had an average speed faster than 9.28s per 100m.
Valpey and Ben D: they measure instantaneous speed in the long jump and in the triple jump, so I guess they could do the same in 100 or 200m, maybe they don’t because in the jumps they know the top speed is achieved just before the jump itself, while in the 100m it may be not achieved in the final line (where the sensors would be placed).
You’re totally missing a key point. Starting the race and getting up to speed becomes a smaller portion of the race with longer distances. This is why anything shorter than 100 meters will never allow for faster average speeds. Of course, your point is well taken that one can only run so fast for so long, so finding the optimal speed would possibly help determine the “world’s fastest” man.
Actually, sprinters achieve a much faster average speed on the 100m from a running start, as in the 4 x 100m relay. Donovan Bailey ran a 100m anchor leg in 8.95 at Atlanta in 1996, and Asafa Powell has clocked a time of 8.84 in the same situation. This is, as far as I know, the fastest average speed ever recorded in an official track meet.
It would seem to me that the average speed would be most affected by the acceleration period at the start of the race. The additional length of the 200 dilutes the impact of this inherently slower period. Thus, assuming the sprinter can maintain his top speed for the duration of the race, the longer races will carry the highest averages.
I’d love to see the speed broke down by 10 or 25 meter segments and look at a graph of speed by distance unit. At what point was speed peak? How quickly is that peak reached? How quickly after the peak is reached does speed decay? And perhaps most germane to the “fastest man” argument, what was the max speed reached and by whom?
## 3 and 4 (and others):
My initial reaction was the same as yours: the “World’s Fastest Man” is naturally he who ran at the fastest speed (i.e. the fastest instantaneous speed regardless of the distance). However, I suspect that a runner’s speed varies slightly depending on where in the stride that runner is. Would you measure it (a) when the heel touches the ground, (b) when the foot is flat on the ground, or (c) any of the infinite other points in a stride? Any such selection would be arbitrary and invite subjectivity. Would it be fair to say one runner is faster just because he happens to be faster in one part of his stride?
I suggest the measurement be the average speed for one discrete stride. This would eliminate the problem above but also would not force the measurer to make some arbitrary determination of distance (100m vs 200m, etc).
Problems with this approach?
Chappy:
My optimal distance is not your optimal distance. Which of us is right?