One switch for the whole portfolio

The definition is an old one. Take the average of NIFTY's closing level over the last twenty trading days, and the average over the last two hundred. When the short average is above the long one, recent prices are higher than the longer past, and the market is in a rising mode. When the short average falls below, it is not.

Two hundred days is roughly ten months, and that length is the point. A twenty-day wobble does not move a ten-month average, so the switch does not flip on a bad week. It flips when the market has genuinely changed character, which happens a few times in a decade. In our five-year test, with ten companies bought and sold together, the strategy made about thirty trades in total, nearly all of them a basket bought or sold together.

Every other share strategy on this site judges each company on its own. This one deliberately does not, and the reason is what large companies do in a falling market: they fall together. When NIFTY drops a fifth, it is not because ten unrelated things happened to ten companies. It is one thing happening to all of them. A rule that judges the market once, and acts on all ten, is answering the question that actually matters in a crash.

The cost of that choice is that it has no view on an individual company. If one of the ten is having a bad year while the market rises, the strategy holds it anyway. The only concession is a stop under each stock, so one that falls hard on its own is sold without waiting for the market to turn.

It is not trying to beat the market while the market is rising. It is trying to be in cash for most of the fall.

Averages are late by construction. By the time the twenty-day average has fallen through the two-hundred-day, the market has already dropped some way, and by the time it climbs back through, the recovery is already under way. The strategy misses the top of every rise and the bottom of every fall. People who dislike this rule usually dislike it for that reason, and they are right about the fact.

What they miss is what the lag buys. The strategy is never fully invested through a long decline. It sells, late, and then sits in cash for the months in which a buy-and-hold investor watches the rest of the fall. That is the one case the rule exists for, and the only one in which it is meant to beat holding. In a sideways year it does the opposite: the averages cross, uncross and cross again, and each false switch costs a little on twenty trades. That is the price, and it is paid in the dull years so that it can be collected in the bad one.

Someone who would otherwise buy an index fund and hold it, and who knows from experience that they cannot actually hold it through a crash. For that person the strategy is a rule that sells before they panic and buys back before they feel brave, which is the opposite of the order most people manage on their own.

It is not for someone who wants to beat the market. In a rising year it earns roughly what the basket earns, less the charges on the way in. The three versions differ only in how much of the account goes into each company, and therefore how much of the account is in the market at all when the switch is on.