Hmm .... several important assumptions in this short post, the most important of which are right here: " ... This isn't as likely to outperform using single company stocks or managed diversified mutual funds ..."
1) In the context of retirement savings, here is what William Bernstein says about stocks: “Do you think that by choosing a portfolio of only a few stocks that you hope will score big, you are maximizing your chances of becoming wealthy? Indeed you are, but you are also maximizing the chances of a retirement of cat food cuisine.” So, for most of us, the casino of individual stock picking is not a wise option.
2) Decades of research show that managed funds in general underperform their benchmarks. Looking over a year, typically one in three will outperform and 7% will go out of business. Looking over an investor's time horizon of five or ten years, the number that outperform is a single digit percentage. Worse, no one has figured out how to predict the outperformance ahead of time.
3) As you conjecture, over the long term "reversion to the mean" tends to move sectors toward their long term average performance but the timing of this is impossible to predict. For example, the US stock market has outperformed the Internationals for about ten years but for the previous similar period it was opposite.
4) If such a simple "outperform the market" strategy actually worked, it would attract tens of billions of dollars of managed money, which would result in the prices in the "down" sectors being bid up and prices being bid down when the sector moved up. This would effectively negate the strategy.
TANSTAAFL, unfortunately. We all wish it was otherwise.