I think you have to look at this comparatively though. The number of rich people is by definition small. To make a comparison worth anything at all is a proble

I think you have to look at this comparatively though. The number of rich people is by definition small. To make a comparison worth anything at all is a problem. While you might or might not like thinking of VCs as a filter, they do perform this function to an extent. If your non-VC sample consists of mostly VC rejects, that's a problem.