We have a good deal of evidence that our selection process is better than random. We know it's at least internally consistent, in the sense that startups that are ranked higher in the application phase are more likely to make it past the interview phase. And we also in turn have (a necessarily small amount of) evidence that the startups that turn out to be big winners do the best in the interview phase.
Are you willing to share what your evidence is that the big winners do the best during the interview phase? Do you rank all startups that are accepted into YCombinator?
Have later YCombinator classes had a larger percentage of homeruns? If you have more applicants and have gotten better at selecting, this should be the case.
Is it about selecting winners, or weeding out losers? Are there companies that you know will fail? Who would be least likely to succeed as an entrepreneur?
Sure; the evidence is pretty low tech. Between us we can remember the interviews of all the most successful startups, and in no case there was any debate about whether we should fund them.
Home runs are so rare that it's not a matter of percentage per batch. A batch will have 1 or 0, and probably 0. It will be a few years before I can tell if the rate is increasing.
Selecting winners and weeding out losers seem the same thing to me. There are companies we think will almost certainly fail, but we can never be sure. An ineffectual person would be least likely to succeed as a startup founder.
