To understand how fast a human can ultimately run, we need to go beyond the record books and understand how Jamaican sprinter Usain Bolt’s legs work.

In 2008, at the Beijing Olympic Games, Usain Bolt ran the 100 m in just 9.69 seconds, setting a new world record. A year later, Usain Bolt surpassed his own feat with an astonishing 9.58-second run at the 2009 Berlin World Championships. With the 2012 Olympic Games set to begin in London, the sporting world hopes Usain Bolt will overcome his recent hamstring problems and lead yet another victorious attack on the sprinting record. He is arguably the fastest man in history, but just how fast could be possibly go?

That’s a surprisingly difficult question to answer, and ploughing through the record books is of little help.

“People have played with the statistical data so much and made so many predictions. I don’t think people who work on mechanics take them very seriously,” says John Hutchinson, who studies how animals move at the Royal Veterinary College in London, UK.

The problem is that the progression of sprinting records is characterized by tortoise-like lulls and hare-like… well… sprints. People are getting faster, but in an unpredictable way. From 1991 to 2007, eight athletes chipped 0.16 seconds off the record. Bolt did the same in just over one year. Before 2008, mathematician Reza Noubary calculated that “the ultimate time for [the] 100 meter dash is 9.44 seconds.” Following Usain Bolt’s Beijing performance, he told Wired that the prediction “would probably go down a little bit”.

John Barrow from the University of Cambridge – another mathematician – has identified three ways in which Usain Bolt could improve his speed: being quicker off the mark; running with a stronger tailwind; and running at higher altitudes where thinner air would exert less drag upon him. These tricks may work, but they’re also somewhat unsatisfying. We really want to know whether flexing muscles and bending joints could send a sprinter over the finish line in 9 seconds, without relying on environmental providence.

To understand how fast a human can ultimately run, we need to go beyond the record books and understand how Jamaican sprinter Usain Bolt's legs work

To answer that, we have to look at the physics of a sprinting leg. And that means running headfirst into a wall of ignorance.

“It’s tougher to get a handle on sprinting mechanics than on feats of strength or endurance,” says Peter Weyand from Southern Methodist University, who has been studying the science of running for decades.

By comparison, Peter Weyand says that we can tweak a cyclist’s weight, position and aerodynamic shape, and predict how that will affect their performance in the Tour de France.

“We know down to 1%, or maybe even smaller, what sort of performance bumps you’ll get,” he says.

“In sprinting, it’s a black hole. You don’t have those sorts of predictive relationships.”

Our ignorance is understandable. By their nature, sprints are very short, so scientists can only make measurements in a limited window of time. On top of that, the factors that govern running speed are anything but intuitive.

Peter Weyand divides each cycle of a runner’s leg into what happens when their foot is in the air, and what happens when it’s on the ground. The former is surprisingly irrelevant. Back in 2000, Peter Weyand showed that, at top speed, every runner takes around a third of a second to pick their foot up and put it down again.

“It’s the same from Usain Bolt to Grandma,” he says.

“She can’t run as fast as him but at her top speed, she’s repositioning her foot at the same speed.”

That third of a second in the air – the swing time – is probably close to a biological limit. Peter Weyand thinks that there is very little that people can do to improve on it, with a notable exception. Oscar Pistorius, the South African double-amputee, runs on artificial carbon-fibre legs that each weigh less than half of what a normal fleshy limb would do. With this lighter load, he can swing his legs around 20% faster than a runner with intact limbs, moving at the same speed.

For most runners though, speed is largely determined by how much force they can apply when their foot is on the ground. They have two simple options for running faster: hit the ground harder, or exert the same force over a longer period.

The second option partly explains why greyhounds and cheetahs are so fast. They maximize their time on the ground using their bendy backbones. As their front feet land, their spines bend and collapse, so their back halves spend more time in the air before they have to come down. Then, their spines decompress, giving their front halves more time in the air and their back legs more time on the ground.

Such tricks aren’t available to us two-legged humans, but technology provides alternatives. In the 1990s, speed skaters started using a new breed of “clap skates” where the blade is hinged to the front of the boot, rather than firmly fixed. As the skaters pushed back, the new design kept their blades in longer contact with the ice, allowing them to exert the same force over more time. Speed records suddenly fell.

People have tried to duplicate the same effect with running shoes, but with little success. That’s because a running leg behaves a bit like a pogo stick. As it hits the ground, it compresses. As it steps off, it gets a bit of elastic rebound. Technologies that try to alter a runner’s gait tend to interfere with this rebound, and diminish the leg’s overall performance.

“It’s hard to intervene in a similar manner to the clap-skates without buggering up the other mechanics of the limb,” says Peter Weyand. (Again, Oscar Pistorius bucks the trend because his artificial legs are springier than natural ones, and give him around 10% longer on the ground than other runners.)

For those with intact limbs, one option remains: exert more force on the ground. Put simply, fast people hit the ground more forcefully than slow people, relative to their body weight. But we know very little about what contributes to that force, and we are terrible at predicting it based on a runner’s physique or movements.

We know that champion male sprinters can hit the ground with a force that’s around 2.5 times their body weight (most people manage around two times). When Usain Bolt’s foot lands, it applies around 900 pounds (400 kg) of force for a few milliseconds, and continues pushing for around 90 more.

Peter Weyand likes to imagine a weightlifter trying to apply the same force in a one-legged squat – they would come nowhere close.

“What we know about force under static conditions under-predicts how hard sprinters hit by a factor of two,” he says.

“We just don’t have the ability to go from the movements of the body to the force on the ground.”

Even if a sprinter’s muscles were eventually boosted by gene doping techniques, we have no way of calculating how much faster their owners would run.

Studies are underway to fill in those gaps, and Peter Weyand is hoping that we’ll be able to make better predictions in five or 10 years. Just a few months ago, Marcus Pandy from the University of Melbourne used computer simulations of sprinters to show that the calf muscles, more than any others, determine the amount of force that runners apply to the ground. At top speeds, the hip muscles become increasingly important too. “Maybe if you train a sprinter, you could potentially train them to have really strong calves,” says Hutchinson.

For the moment, however, any predictions about the ceilings of human speed are still ill-informed ones. The only way to work out if Usain Bolt or some other sprinter will smash the existing record is to watch them.