I completely agree with that - the goal of any thesis should be to prove the validity of a design implementation, and in this case that should include hard medical evidence.
I am only trying to convey a spirit of simplicity - do not try to solve everything but starting simple, focusing on the most interesting aspect and developing a refreshing design implementation. Look for example at the SmartGoals project - www.smartgoals.nl. It is extremely simple, interconnected cones with flashing lights. And it does not cover many aspects of the game - but it is a great tool for ball passing, teamwork and insight.