And obviously, success is higher when selecting a strategy more likely to succeed.

For my money I'm going to expose the person to a better strategy rather than supporting and reinforcing a clearly bad one.

+1. Using statistics about a population to make determinations about a specific outcome is often fallacious anyway. “70% of businesses fail” might sound like it is a bad idea to start one, but of course every business succeeds or fails on its own specific merits.