The most important thing in terms of your circle of competence is not how large the area of it is, but how well you've defined the perimeter.

Investors should be skeptical of history-based models. Constructed by a nerdy-sounding priesthood using esoteric terms such as beta, gamma, sigma and the like, these models tend to look impressive. Too often, though, investors forget to examine the assumptions behind the models. Beware of geeks bearing formulas.

What we learn from history is that people don't learn from history.

You do things when the opportunities come along. I've had periods in my life when I've had a bundle of ideas come along, and I've had long dry spells. If I get an idea next week, I'll do something. If not, I won't do a damn thing.

We've used derivatives for many, many years. I don't think derivatives are evil, per se, I think they are dangerous.

I always say that in investing you want to buy stock in a company that has a business that's so good that an idiot can run it, because sooner or later one will. We have a country like that.

I wouldn’t mind going to jail if I had three cellmates who played bridge