Read my review

How I Use Darvas Box Theory

I try to follow the system as closely as possible, but I find that I sometimes have to make changes to the Darvas rules just a bit.

Surging Volume

Darvas used surging volume as an indication that a stock was worthy of watching. I don't do this, simply because I haven't figured out a good way of doing it mechanically. Darvas himself used Barron's weekly data to identify surging volume, which is fine for his time. However, in this day and age, I think doing anything manually would be too time consuming and wasteful. When (if) I get around to writing a program to automate the screen, I might use it. Or maybe there is some tool out there (that is cheap) that I can use...not sure.
In lieu of surging volume, I use Investor's Business Daily (IBD) as a proxy. The theory is that only strong stocks make it into the IBD lists; therefore, if I use their lists I will be using strong stocks, regardless of volume contribution. It is possible that getting on the list accompanies a surge in volume, or it could be the other way: get on the list first, then a surge in volume. It is worth noting that Investor's Business Daily's data is used by institutions, so ith might not be so far fetched to see a volume surge after a stock gets on one of their lists.

What I do is look at the newspaper each Friday night (after they update the IBD 100), and find the five industries that are currently the strongest in relative strength. Then I use their 197 Industry Groups to find stocks with high ratings (90 or higher). I think this solves Darvas problem of finding stocks that are fundamentally strong (with good earnings prospects and strong uptrend in earnings growth).

Alternatively, one could also use the IBD 100 directly as a proxy for strong earnings. This is okay too. My only concern is that because stocks generally following their industry (and the market as a whole), if you have a strong stock in a crappy industry, you're fighting a headwind unnecessarily.

All-Time vs. 52-Week High

I don't think Darvas ever said whether he used all-time or 52-week high. He only said that an all-time high needed to be known, as well as highs for the past two or three years. Personally, I prefer to use the all-time high. However, in exceptionally strong cases, I will settle for a 52-week high. It's just a personal preference - I haven't done a back test to determine if one or the other is better in terms of winning percentage.

Box Formation

In this, I differ from Darvas in that I use weeks, not days. Other than that, I follow Darvas' stipulated methodology. Otherwise, I guess it wouldn't really be the Darvas method...? The sticking point is whether or not the three-day (week) rule includes the day the high (low) is made. In my experience, I find that since I often do not see the boxes developing, it doesn't really matter to the upside. As a practical matter I prefer having more weeks of consolidation to fewer, so I do not include the high day (week) in the three days count. Thus, including the high day there are four days (the day itself, plus three days of no penetration. To the downside I also do not include the week in which the low was made.


When it comes time to enter, I use, as Darvas did, daily quotes. In my case I use real-time daily quotes, since I cannot use pre-set stop or limit orders. I set the system to send me a text message when the stock trips the buy point, then I either buy a call option on the stock, or use some sort of options strategy on it. Which I use depends on how I think the stock will perform in the short run, and how much cash I have available. Generally, I try to buy cheap OTM options. Failing that, I do a spread to lower the cost and break even point. If I am not confident of the market's direction or the stock's performance, I buy a straddle as cheaply as possible.


When I exit depends on what happens to the stock. If it breaks to the downside, I will close the position. If it moves up, I either close it out (spreads), or le it ride (straight).

Pyramiding (Laddering)

I try to do this as often as I can since when the going is good the profits really pile up. However, I have found that I lack the discipline to make good use of pyramiding.
Page: 1 2 3 4