3. Microscopically increase the success of onside kicks.
Onside kicks work about 24 percent of the time. But that's a misleading figure; they rarely work if the opponent knows they're coming. From an entertainment perspective, it would be nice to increase the likelihood of the kicking team recovering its onside attempt, but not by too much—if teams suddenly recovered 40 percent, the game would become a farce. So how do we slightly increase that possibility? By changing where the receiving team lines up. Right now, it's ten yards from where the ball is kicked. Make that fifteen yards, which would force the receiving team to charge forward once the ball is kicked. It wouldn't change the metrics dramatically, but it would give the trailing team slightly better odds.
While I agree with him that this would probably make onside kicks more challenging for the teams involved, I think it would also drastically increase injuries, since you've now introduced the element of charging athletes to both sides of the ball. I can't imagine their collisions wouldn't cause some significant damage. And in this world of concussions under the microscope, I think we might want to rethink this one.