There is no disputing any time you convert one energy to another you loose efficientcy.
If you can capture energy like water flowing thru a dam and IMHO wheels that are spinning anyway seems to me you would benefit.
I keep hearing about the amount of friction that has to be overcome by an alternator at high output...the only time there would be a high output is when you are using the stored up energy you hopefully built up it the 100 miles you just ran on gas and for a short time if you ran the batteries way down.
I am not proposing to take fire from the magneto I am talking about an alternator's pulley resting on top of your tire.
Someone said maybe my math was off here's my formula for approximate rpm: 1 mile = 5280 feet I am assuming an alternator pulley is less than 1 foot in circumference. Your tires be they trailer 10 inch or bike 26 inch still cover the same distance.
Surely the friction of my 245 pounds on the tires grabbing at the pavement makes that alternator pale in comparison.
Have ya ever seen how hard the guy pulls down to test a load on an alternator at the auto parts store?
So if my numbers are correct 5280 feet times let's say average 20 MPH equals 105600 feet of distance divided by 60 minutes in an hour equals 1760 feet per minute hence 1760 RPM which is enough to produce nominal amps and volts from the alternator.
What would be great is if you could put a controller that would increase the amps load output from the alternator to increase if you went over say 25 MPH as you would be surely coasting downhill and restrict the load output to zero when engine rpm is low.
Here's someone who kicked this idea around in a race car...he had a switch to turn off connection to alternator during 1/4 mile race:
Switching off alternator during hard acceleration Text - Physics Forums Library
I was really just doing the armchair engineering thing but I think I'm gonna try this some day when I have more funds...unemployed and obviously boored at the time now.