My addition to the Third Option would be: if you know something's a good company, wait until a cyclical (but fundamentally extraneous to the company's business prospects) market downturn and buy it while everything is crashing.

I think this is basically wrong, because opportunities are time-sensitive. If a company is undervalued now, it's not obvious it will remain undervalued until the next cyclical downturn, and you pass up on the benefits of any market correction in the valuation of the undervalued company.

I do agree that it makes sense to invest countercyclically (where you have more of your wealth in stocks when you think the stock market is undervalued, and more of your wealth in cash / CDs / etc. when you think the stock market is overvalued), but determining whether the stock market as a whole is undervalued or overvalued is a difficult task, and it takes planning and forethought to ensure you are not cash crunched when the economy dips (which you should do now).

I also think that correctly pricing downturn risks is a subset of correctly pricing shocks in general. How much damage will the oil spill actually do to BP? How much damage will Jobs's death do to Apple? How much damage will Buffet's death do to Berkshire Hathaway? How much damage will a general economic downturn do to Apple?

I'm pessimistic on Apple's prospects without Jobs, because of what I know about his management style, but time will tell how that turns out. I'm optimistic about BRK's prospects without Buffet, again because of what I know about his management style--and so if the market dips significantly when they take his pulse again, I'll buy BRK (like I bought BP when the market overestimated the damage). And here we're in the same sort of situation- if you think that BRK is will grow in both the short-term and long-term, but there's an upcoming predictable dip (Buffet's death), do you wait for the predictable dip to buy, just buy now, or split some funds out to buy now and other funds to wait for the dip?