My view is that an investor is better off knowing a lot about a few investments than knowing a little about each of a great many holdings. One's very best idea's are likely to generate higher returns for a given level of risk than one's hundredth or thousandth best idea.

Most institutional investors feel compelled to swing at almost every pitch and forgo batting selectivity for frequency.

Ratings agencies are highly conflicted, unimaginative dupes. They are blissfully unaware of adverse selection and moral hazard. Investors should never trust them.

Investors should pay attention not only to whether but also to why current holdings are undervalued. It is critical to know why you have made an investment and to sell when the reason for owning it no longer applies. Look for investments with catalysts that may assist directly in the realization of underlying value. Give reference to companies having good managements with a personal financial stake in the business. Finally, diversify your holdings and hedge when it is financially attractive to do so.

Be sure that you are well compensated for illiquidity - especially illiquidity without control - because it can create particularly high opportunity costs.

In a world in which most investors appear interested in figuring out how to make money every second and chase the idea du jour, there's also something validating about the message that it's okay to do nothing and wait for opportunities to present themselves or to pay off. That's lonely and contrary a lot of the time, but reminding yourself that that's what it takes is quite helpful.

The government - the ultimate short-term-oriented player - cannot withstand much pain in the economy or the financial markets. Bailouts and rescues are likely to occur, though not with sufficient predictability for investors to comfortably take advantage. The government will take enormous risks in such interventions, especially if the expenses can be conveniently deferred to the future. Some of the price-tag is in the form of back- stops and guarantees, whose cost is almost impossible to determine.