You have to make decisions with incomplete and unscientific data all the time. If there was a scientific predictor to business success, VCs wouldn't pick lemon

You have to make decisions with incomplete and unscientific data all the time. If there was a scientific predictor to business success, VCs wouldn't pick lemons ~80% of the time. There's nothing wrong with making decisions based on features like that, as long as you're accurately measuring your results. Who knows, maybe his VC friend is "retired" for a reason?