Some technical ideas are stupid in a trolly way: it's difficult to actually put numbers to the reason this is bad, because you have not actually put forth a pro

Some technical ideas are stupid in a trolly way: it's difficult to actually put numbers to the reason this is bad, because you have not actually put forth a proposal, only an idea, and we would require a detailed physical simulation of your proposal to be able to demonstrate with proof "This is a fucking stupid idea". Nevertheless, design enough things and mess around with energy capture schemes enough and you do develop some intuition in the matter of how hard it would be to implement, how much you would actually get back, how negligible that is. That intuition is screaming at us that this is a waste of time, your idea is bad, and you should feel bad. What is easy to point out that you're trying to shift the burden for electrical power generation from solar/nuclear/coal, on to scarce and expensive oil, mediated through low efficiency automotive engines, and in the process increase road construction costs by orders of magnitude. The only free lunch you get in automotive regenerative power for every car that passes, is on the downslope of very long hills at high speed. You can regenerate power here through the wheels (and that's slowly being put into the fleet as we start to use wheelhub motors & large battery banks), or through in-road baffles. You could do the same thing at traffic lights, but they don't have a predictable deceleration curve and speeds are much lower (making it feel choppy) and it's not an environment where you want to mess with people's sense of acceleration, for safety reasons.