All told, results have been mediocre in the last 25 years, I think, since the GenRe acquisition. Berkshire has still done remarkably well, roughly keeping up with the S&P, probably because it has avoided really big catastrophes, and because the Apple success has pulled up results a lot, and because of LEVERAGE.
So I can't prove it, but I am sticking with my hypothesis that ok results with good leverage have been the key to success, at least for the last 25 years.

The quality that many people miss from Mr Buffett's stock picking skill is not that that the companies he picks do amazingly better than the market, the thing that so many people try to look for, but rather that he tends not to put big money into things that go wrong. He has never had a realized loss of 1% of any portfolio he has managed.

It's not a perfect comparison, because there is no pre-known holding period for Berkshire's investments, but just for comparison, since 1997 a random S&P 500 company held for 3 years has a 32.5% chance of losing money if held for 3 years (including dividends and before tax), and the average loss of those positions is -34.5%. (the average winner is up 68.3%).

People usually focus on huge winners. But they aren't really needed. It's amazing how much better one's portfolio does if one merely skips a significant fraction of the big losers.

Rule #1 works, if you can manage it.

Jim