It has been suggested that any marginal benefit (for some people) of ‘evening out’ pedal effort would be better if long/short sides were in a different orientation from the one Shimano decided.
Sheldon was always happy to fly the flag for Biopace, and offered a pretty good explanation of why Shimano put the smaller radius at the power stroke portion and the larger radius at the deadspots.
Oval rings were a marketing gimmick.
No-one really ran clipless pedals before Look thought of applying ski binding technology to bike pedals, and went all out to sponsor riders in major races. Biopace, and its imitators, probably was a technological solution to a problem that no-one had – or at least, one to which no-one had thought to apply the finest Japanese scientific minds. The crank and round chainring is already a very biomechanically efficient system, because it provides roughly a 1:5 power to recovery ratio for muscles, and using more of the rotation to generate useful power will affect stamina, just as using our 'useless' arm muscles to help propel a bike in tandem with (fully functioning) legs is not a way to go faster for longer, when the leg muscles are already capable of using all the energy the body is able to generate.
I came to mountain bikes just when Biopace was hot stuff, but by the time I had graduated from my Sachs geary Raleigh made of weighs-a-ton steel to my Suntour XPress crmo Trek, oval chainrings were already passé, on the basis that round chainrings were 'more powerful' and probably better after all. I toy with the idea of putting a vintage Biopace chainset on my blue bike, but it's from 1991 and thus a couple of years too young for proper retrobike approval.