Read my review

Details Behind the Darvas Box Theory

As always, nothing is as quick or easy as it seems. What follows below is a dissection of the various details that are required when trying to find a good candidate to buy and when trying to identify Darvas boxes.


Even though Darvas' method is primarily a technical indicator, Darvas himself says he married a fundamental analysis approach with a technical analysis approach when it came to actually buying the stock. In his book, he talked about this idea of selecting only stocks that were in new and growing industries, and had a 20-year timeframe for when these new products would come to dominate their respective industries. During his time, WW II just ended not long ago, so he saw industries such as electronics, missiles, and rocket fuel as bing "tied up with the future". According to Darvas, it all boiled down to one thing: "improving earnings power or anticipationg of it". That's about as fundamental as Darvas gets in his book.


On the technical side, there are quite a few details to be hammered out. First, it appears that Darvas used an unusual surge in volume as an indicator. He continuously refers to seeing unusually high volume in Barron's as his reason for keeping tabs on a stock. Therefore, we can assume that is the first of his rules: Keep an eye on stocks with unusually high volume activity. Throughout the latter part of the book, he refers to stocks that rise like a beacon out of the baby bear market.

Second, it is unclear whether Darvas used an all-time high or a 52-week high as a requirement for the stocks he would consider. I am not sure he mentions it explicitly in his book, and looking at the charts in the back don't really help either. In practice, I have found that either is good, but all-time high is better.

Finally, there is this little problem of identifying a Darvas box. In the book, Darvas himself never actually mentions this particular detail. I am not sure if he was being vague on purpose, nor am I certain that he doesn't explain it in some other book (or some later edition of this one). He does, however, give details in the appendix of my copy (he says those questions were frequently asked, but reading some of them you have to wonder about that).

The following is based on my reading of the appendix. To identify a Darvas box, one would first locate the highest price that a stock reached during a certain period. Then, if that price is not breached to the upside for three days, Darvas would consider it to be the top of the box. It is only after locating the top of the box that one begins to search for the bottom. Darvas was adament about this point. Locating the bottom of the box is essentially the same as looking for the top, but in reverse. Once the box has been defined, says Darvas, the stock should bounce around in it. If it does not, he would be worried. If, at some point in the future, the stock pierces the top of the established box ('by a tick' says Darvas), then one would immediately buy the stock.

In the book, Darvas does not metnion whether or not the 'three days' includes the day in which the top of the box occurs. I am not really sure. And another thing: Darvas uses weekly Barron's data to find his stock candidates, but uses daily quotes to locate his boxes. I depart from this slightly: I always use weekly charts, even to find breakouts. In my experience, I have found that this cuts down on the number of false breakouts.

Over the years, I have seen all manner of Darvas boxes take shape. Notice that Darvas himself does not stipulate as to what the internal structure of the box needs to look like. He only identifies the dimensions and lets the stock run its course. I have found that, in practice, stocks that bounce around in relatively straight, horizontal fashion perform quite well, as do stocks that have a symmetrical wave shape.

Page: 1 2 3 4