Bear in mind that an alkene happens to be electron deficient. The argument of hyperconjugation is that the sigma bond of one substituent on the double bond donates to the pi* orbital as you have mentioned. That allows the electron deficient carbon to 'share' in some of the electron density of the substituent. It's a poor resonance structure to be sure, but it's another resonance structure, and, as someone mentioned, is an explanation of an observed experiment. It is a little fishy. Perhaps this is a bit oversimplified.
Also, not all more highly substituted alkenes are more electron rich. It depends on the groups. Take a look at tetrafluoroethylene, for instance. That's certainly less electron rich than ethylene. This is, of course, speaking of the double bonds, not about the number of electrons in the total molecule. Yes, TFE has more electons than ethylene, but the double bond in TFE is much more electrophilic.