> You'd intuitively assume that agents become more rational the farther up the ladder you go, but this seems like a testament to the contrary.
This is a well known bet from Warren Buffett, but you should not take conclusions without really understanding what the bet is about, and its underlying assumptions.
- The first assumption is that you're in for 10 years, this is the duration of the bet.
- The second assumption is that you don't need that money during the whole duration of the bet.
Why are these assumptions important? Because different investors have different risk profiles.
If you are in your twenties, saving for your retirement, then you should definitely follow Buffett's advice: just "buy the market" (S&P, Russel, whatever low expense ratio ETF or index fund tracking these will do).
This is because the longer horizon you have on an investment, the higher the volatility you can afford. It will always average out after a while.
Imagine now you have a different risk profile:
- You are now in your late fifties and expect to retire in ~5 years;
- You are an insurance company and may have to withdraw from your portfolio at any time to cover for expenses of your clients;
- You are an income investor and expect to live from the interests of your portfolio;
Would you be well advised to invest in an equity index? Definitely no! The volatility will crush you at the worst time, and you cannot afford to wait 5/6 years for the market to catchup.
In these scenarios (and there are plenty of them) what you want is diversification. You can accept to have less performance, in exchange for less volatility; so that once you sum all your different, uncorrelated portfolios, you have something with the risk profile that you can afford.
On a more local scale, you can apply this idea intra-portfolio. This is called "minimum variance" optimization.
Now there are a lot of things you can do to tailor your portfolio to your preferred risk profile. Traditional literature (e.g. "The Smart Investor") will advise you to buy some investment grade bonds, since they are almost anti-correlated to equities. You could also buy some real-estate, that will provide you cover during recession or very high inflation. etc, etc, etc.
What I want to emphasize here is that you cannot just look at the performance of a fund and decide whether it's a good or bad investment. If it exists, it means someone, somewhere, has a need for such investment, because it fits well with its current portfolio.
> He explained that he understood that any strategy was likely to have runs of poor performance, and his idea whenever it happened was to "swap" to a better performing strategy which was having a good run at the same time.
That's called "market timing", it's the unicorn of alpha combination. You will find multiple papers on internet explaining you why this is incredibly hard to do, and "beating the 1/n" is almost never possible.