Hi Guys,
Interesting article and interesting to hear the arguments. I run a blog as part of my Research group and thought this would be a good subject for a topic (I'm a sports Engineer) so have done some quick calcs with some simple assumptions. I'll write it up this week and get it on ASAP (there's a good article waiting to go up so can't push in front) but to get a sneak peek on the numbers.
Assuming a rider + gear weight of 80kg, total bike weight of 12.7 kg and wheels weight (in tyres etc. of 3.12 kg) by decreasing the rim mass by 100 g (and putting the mass elsewhere on the bike) you get a 4% increase in acceleration for a zero change in the bike mass.
Interesting stuff, I'll write it up properly and put in some real examples (i.e. how does it relate to time, given an equal torque at the wheel)
Also, realise this is a very simple assumed system, accleration is exactly that, accln of a point mass with two rotating bodies depending on a torque applied at on a rotating mass.
Hope that's of some help to the OP. I can't answer whether that 4% is noticeable though!
Si C