Mark Hulbert had a piece for Barron's that says the 200 day moving average (DMA) is no longer a useful market timing tool noting that there have been many false positives. Here's the chart he used to show examples.
I have no pushback on this point, the indicator is prone to false positives. I have framed how to use the 200 DMA as an effective tool for many years and the points made by Hulbert, not being new, are ones I have taken into account.
First and foremost a portfolio is a series of decisions, any of which could be wrong including when the S&P 500 breaches its 200 DMA. The possibility of being wrong is why when I do take defensive action I start slowly. Back in May I bought AGFiQ US Market Neutral Anti-Beta (BTAL) for clients after selling their Square (SQ) as a first defensive step thinking that the 2% rule had been invoked. Square had gone from $40 to $70 in a few months and so it seemed prudent to sell, obviously it is gone a lot higher. When I wrote about BTAL I said I chose that instead of an inverse fund in case the market was not rolling over, I thought BTAL would be less of a drag versus something like ProShares Short S&P 500 (SH), I was less sure about the market rolling over but my process called for defensive action. Since the trade, BTAL is up almost 5% while SH is down about 10%.
The drag from this has been small but I stuck to my process, discipline is crucial.
The reason I like the 200 DMA as an indicator is that it is simple. When an index breaches its 200 DMA it indicates that there is some sort of problem with demand for equities. That problem may be serious or not so serious but it is a problem. The chart above shows that most of the problems turned out to not be serious. There is no way to be certain on day one or two after a breach whether the problem is serious (although there might be some hints related to the slope of the moving average and a couple of other things). What is (almost) certain is that a serious problem will play out over several months. Bear markets start slowly, typically giving several months to get out. Crashes are different, they tend to snap most of the way back quickly, they have tended to be better to buy than to sell. Even if you can't buy in the face of a crash, at the very least, not selling can be a difference maker.
When you think of a breach of the 200 DMA as a first warning and take only small actions then the utility starts to make more sense. Getting completely out of stocks or completely in based on one indicator is guessing which is a lousy strategy. The point I make above is crucial, bear markets give you months to get out. In both of the last two, the S&P 500 was only down about 10% six months after the peak. As markets slowly deteriorate I take more defensive action. I believe BTAL can go up when the next bear market hits (it sells short high beta and goes long low beta). Inverse funds should go up in bear markets (they're designed to do this and I believe they can generally get it done the next time). Gold has a historical tendency to have a negative correlation to equities, I own gold for clients too. If these three (BTAL, SH and gold) do go up in the next bear market they will grow to hedge more of the portfolio as the market drops. As this plays out it would only take a few sales in addition to those three diversifiers to protect the portfolio and by protect I simply mean avoid the full brunt of a large decline.
So the next breach is either the real thing or it isn't. You don't need to be concerned with making a correct proclamation that a bear has started, you probably just care about trying to protect your portfolio. This affords being able to take defensive action slowly. This protects against being wrong (you still have most of your equity exposure at this point) and it protects against the real thing (you have initiated defense).