Originally Posted by Mothra
One of the reasons you can exert more force than your body-weight is that you'll pulling up on one side while pushing down on the other. The back-muscles is also used when pulling up on the bars to counteract the upward movement of the body. So you won't automaticaly launch yourself off the bike if you apply more than your body-weight. Of course, it takes years, decades of training in order to get the brain/muscle-coordination to spin smoothly while sprinting.
The other thing you have to consider is the gearing. It's really torque at the
rear wheels that makes the bike move. Actually
thrust at the contact patch is what counts. So take torque at the crank, run it through the gear-ratios, divide out radius of rear-wheel and you get linear thrust at the contact patch. You'll notice that the bike's gearing divides (reduces) the torque, while a motorcycle's gearbox multiplies (increases) the torque from the crank.
Finally, it's
power that's the real determination of speed. Which is
torque X RPM, so the faster you spin, the more power you make. A motorcycle's torque multiplied by 9000rpms is gonna generate a heck of a lot more power than a human at 90rpms.
You can take McEwen's 1700-watt output and work backwards through the gearing & RPMs to arrive at his torque@crank and realise the guy's a monster!
+1
My guess is that the torque numbers from the pros isn't all that much higher(and may be less) than that of the general public. But, since they can maintain high rpms all day, they have much better total power. What I think would be a real interesting experiment would be to see how a racer would perform if they carried/pulled a significant load vs someone who rides loaded as a matter of routine, or a real in-shape clydesdale(250 lb with low bodyfat). My guess(and again, it's a guess) is that as the pulled load increases, the clyde would perform better in comparison, as maximum torque becomes more important than maximum power.