Subject: Re: Diversifying away from Berkshire
One random thought:

It is *extraordinarily* difficult to pick a stock that will be a long run outperformer with high reliability.
Including reliability of the future outperformance, reliability of the firm (Rule #1), and reliability of the conclusion.
I no longer think I can do it.

For that reason, there are no easy substitutes for Berkshire as large capilal allocation choices.

Sure, I think there is a good case to be made for a few firms. But doesn't every firm have a case for it?

Perhaps the best way to have another leg on the stool is to add a shotgun approach rather than another 'rifle' pick or few picks.

I mean the sort of thing I suggest from time to time: Pick a random slate of 20-30 companies with good statistical properties.
One of my favourite combinations is high ROE and decent sales growth.
Replace each position every 1-3 years if it no longer meets the cutoff criteria you used at the beginning.
(periodically reconstituting the portfolio is critically important for 'shotgun' investing, for long boring reasons)

A typical strategy might go like this:
* Start with the list of Value Line stocks. They are mostly decently big and plausible picks.
* Of those, find the 40 with the highest reported [recent annual] return on shareholders' equity
* Of those, buy equal dollar amounts of the 20 stocks with the highest reported 5 year rate of growth of revenue per share
* Hold for a while, maybe 1-12 months. 6 is good.
* Repeat. Replace any stocks that no longer meet the same criteria.

Though it will never work as well as it did in a test, reconstituting every 6 months, this would have beat the S&P by 7-8% a year 1986 to date.
If you managed a 2%/year advantage in real life over a long period, it would be well worth the effort.

Not that this is the magic solution. The idea is merely that achieving the level of
(a) very low risk of large permanent capital loss with
(b) quite high chance of outperforming a cap weight index and extremely low chance of material underperformance
...is perhaps (perhaps) best accomplished with a slate of firms, not a few hand picked ones.

I try to remind myself: if you look in the mirror and don't see Warren Buffett, don't invest as if you did.

Jim