Every now and then, the typical LR poster comes out with a hypothesis and it is alwasy predictably along these patterns;
(1) Runner (insert name here) could have won if he had not run in the 2nd lane.
(2) In (insert year probably before 60% of the people on the planet were born ) runner x.McNoun could have run this time (less than current world record).
(3) Based on runner x.McNoun's splits he could have run time xxxx (less than current world record) if he had paced himself properly.
All this are stupid arguments folks, Anyone who ever took a basic statistics class and understood statistics, and this is all statistics folks,it is all statistics, knows one thing, you do not extrapolate beyond or below you data points.
A curve is not linear forever, there are diminishing returns, there are saturation points where more input does not result in more output and so on and so forth. But again let not proper data analysis and sound scientific argument interfere with your fantasies.