But it was the main reason you or I even discuss him right now, which implies that it was one hell of a multiplier.
My problem is that the ratios are very distorted: 1% work vs. 99% luck. Work is necessary but not sufficient is all I'm saying.
Hence how would one optimise the multiplier (whilst obviously working on being just plain good), if that is at all possible.