The Darvas Box Trading System by Nicolas Darvas

This is a book review of two books, both written by Nicolas Darvas. This happens to be the first book review which covers two books at once, and the reason simply is that both books need to be read together because one book makes reference to the other book, and if you don’t read both, you would probably be a little puzzled by some of what the author was writing.

The two books are

1. How I Made $2,000,000 in the Stock Market (1994, first published 1960); and
2. Wall Street: The Other Las Vegas (2008, first published 1964)

Darvas was a full-time professional dancer who travelled around the world with his partner to, well, dance. He did his trading mostly through cablegrams. He asked his brokers to send to him daily quotes on stocks he was interested in and weekly Barron’s stock tables. He then sends instructions to his brokers with his on-stop buy orders and corresponding stop loss orders. He managed to turn around $50,000 to more than $2,000,000. He started trading in 1953, started using his Box Theory in 1956, achieved $2,000,000 by the end of 1959.

The first book, “How I Made $2,000,000 in the Stock Market” is a very good read for the story. In terms of the exact rules of the box, you would get that from “Wall Street: The Other Las Vegas”, with clarifications in a Q&A section at the back of the “How I Made $2,000,000 in the Stock Market”. The story bit in “Wall Street: The Other Las Vegas” is much worse than the first book so I would recommend that you read the first book followed by the second book, then going back to the Q&A section in the first book.

Darvas later had a few other books, including “You Can Still Make It In The Market” where he applied his method through the bear market in the 1970s, and “Darvas System for Over the Counter Profits“, but I have not had the chance to read those [I have just read 1st of these two books where there were some updates in the system rules. You can find the book review here: Book Review of You Can Still Make It In The Market by Nicolas Darvas.].

Darvas’ system is basically one that identifies support and resistance levels using mechanical rules, and buys into stocks trading at all-time highs with increased volume and breaking resistance levels, selling when they drop below support levels. The system is simple, clear, and easily implementable. One thing to note though, it is best to learn straight from the source (i.e. read the books), because some of the materials on the net may not present the Darvas method accurately.

Difference Between Darvas’ and Livermore’s Methods

Note that Darvas’ original system works best in a bull market where there are indeed a number of stocks making all-time highs. It was also not designed to be used to play the short side in a downtrend.

In contrast, Livermore goes for the leading stocks in the leading sectors, in the direction of the overall market trend. Jesse Livermore’s method is meant to be applied to both bull and bear markets, on the most active stocks in the strongest industries. Support and resistance (Darvas calls them the bottom and top of the box, while Livermore calls them pivotal points) are used to identify trend turning points.

Darvas’ method uses support and resistance to identify breakouts of stocks that have something special going on with them (i.e. sudden investor interest). The target stocks are different, and the trading approach taken is also different.

Overall, both books are definitely recommended! They are short books which can be read quickly, the idea is straightforward, and the story in the first book is great.

Key Points

Darvas’ Learning Journey

1. Canadian period
   • Lost money gambling in the Canadian market from tips, rumours, and nice sounding company names.
   • Lost money following financial advisory services and brokers’ tips.
2. New York period
   • Read up on investment books, daily papers, financial columns, subscribed to stock market services (Moody’s, Fitch, S&P).
   • Pored through financial statements of companies, analyzed their fundamentals.
   • Lost money on stocks with supposedly very good fundamentals.
   • Lost money picking stocks based on things like top quality ratings, selling below book value, strong cash position, never cut their dividends, stocks experts like, price earnings ratio, profit margins, industry outlook, choosing the strongest company within the strongest industry group, researching comparable companies, comparing earnings growth of different industry groups, etc.
   • Lost money selling too early for little gains and losses from jumping in and out of losing stocks.
   • Lost money trading in the OTC market due to the huge bid-ask spreads.
   • Lost money listening to more broker’s tips.
   • Lost money following insiders. Insiders are human too, they often bought too late or sold too soon. They might know all about their company but they did not know about the attitude of the market which their stock is sold.
   • Finally got saved with a stock that he bought simply because it was rising (Texas Gulf Producing).
3. [My note]
   • While it seemed like all the things that Darvas tried did not work, I am not convinced that Darvas did some of the methods correctly.
   • The fundamental research that Darvas did seemed too superficial and did not go in-depth into the analysis of critical points such as business model, competitive dynamics, return on invested capital, free cash flow, etc.
   • Similarly, Jesse Livermore did well with strongest stock in the strongest industry, but when Darvas tried it, he picked the cheapest stock in the strongest industry (Jones & Laughlin) even though he called it the strongest stock, probably because he thought the “best bargain” is the strongest.

Darvas’ Discoveries

• Stocks trend
  • I started to realize that stock movements were not completely haphazard. Stocks did not fly like balloons in any direction. As if attracted by a magnet, they had a defined upward or downward trend which once established, tended to continue.
  • … contrary to my impression, stocks behaved with a certain consistency, following upward or downward trends, which made it possible to foresee what they were likely to do on the basis of what they actually were doing.
  • I saw reasons for this; buying tended to generate more buying and progressively higher prices. Conversely, selling at a given price soon exhausts the supply of buyers available at that price, and forces offerings at a lower price — and this process, too, tends to be progressive.
• Stocks move from Box to Box
  • Within this trend stocks moved in a series of frames, or what I began to call “boxes”.
  • They would oscillate fairly consistently between a low and a high point. The area which enclosed the up-and-down movement represented the box or frame.
  • … the entire upward progress of the stock in question had consisted of movements of a similar sort – a progression from one box to another, each stage in the development of the stock being marked by a period of oscillation between clearly discernible limits, then a breakthrough, and a new period of bouncing up and down in the next box, and so on.
• Stocks in an industry group move together
  • It seemed to me, listening to such reports and studying them in the Wall Street Journal, that this was the way the market moved, first one group gaining, then another pushing to the top, like schools of dolphins surfacing one at a time.
• An edge is required
  • There is no sure thing in the market — I was bound to be wrong half of the time.
  • I knew that being right half of the time was not the answer to success. I began to understand how I could break even and still go broke. There was only one answer to this danger: My profits had to  be bigger than my losses.
• Stop loss is required to stop big losses and keep profits
  • The task was to define the frame exactly and be sure the stock did not move decisively below the lower edge of the box. If it did, I sold it at once, because it was not acting right.
  • I decided to give “on stop” orders to buy at a certain figure with an automatic “stop-loss” order on them in case the stock went down. This way, I figured, I would never sleep with a loss. If any of my stocks went below the price I thought they should, I would not own them when I went to bed that night.
  • I knew that many times I would be “stopped out” for the sake of a point just to see my stock climb up immediately after. But I realized that this was not so important as stopping the big losses.
  • I always sold too quickly because I am a coward [afraid that the gains will disappear]. I decided that since I could not train myself not to get scared every time, it was better to adopt another method. This was to hold on to a rising stock but, at the same time, keep raising my stop-loss order parallel with its rise.
• Stop loss foretells market turning points
  • If you find your automatic stop loss selling you out of everything, and the stocks you are watching are not moving towards higher boxes, then it is a signal that the market has changed.
• Companies with rising earnings trends are less affected in any market downturn.
  • I would select stocks on their technical action in the market, but I would only buy them when I could give improving earning power as my fundamental reason for doing so. [Darvas’ techno-fundamentalist theory]
  • … I looked out for those stocks which were tied up with the future and where I could expect that revolutionary new products would sharply improve the company’s earnings.
  • But I did want to know whether the company belonged to a new vigorous infant industry and whether it behaved in the market according to my requirements.
  • … I was constantly searching for the stocks that would climb into the stratosphere because of the vision of their future.
• Buy more as the stock goes higher
  • Darvas usually built his position over 3-5 purchases.
  • The first purchase usually is a smaller purchase, followed by the rest of the purchases roughly of equal size (in terms of number of shares).
• Volatility within the Box is good
  • While it stayed within its box, I considered a reaction from 55 to 50 as quite normal. It did not mean to me that the stock was going to fall back. Just the contrary.
  • It shakes out the weak and frightened stock holders who mistake this reaction for a drop, and enables the stock to advance more rapidly.
• Limit your losses by both amount and time
  • By amount: I have discovered no loss-free Nirvana. But I have been able to limit my losses, without compromise, to less than 10% whereever possible.
  • By time: Profits are a function of time, and so good reasons have to exist to keep a profitless purchase longer than 3 weeks.
• Watch the Market
  • From then on, I made up my mind to keep watching the Dow-Jones Industrial Average, but only in order to determine whether I was in a strong or a weak market.
  • This I did because I realized that a general market cycle influences almost every stock. The main cycles like a bear or a bull market usually creep into the majority of them.
  • By my practice, I found that my methods were best suited to the sort of market in which the greatest opportunity for selection over a wide range of rapidly rising stocks existed — and that, of course, meant a strong bull market.
• Keep a trade journal
  • Whenever I bought a stock, I wrote down my reason for doing so [and the price]. I did the same when I sold it.
  • Whenever a trade ended with a loss, I wrote down the reason I thought caused it. Then I tried not to repeat the same mistake.
• Stick with friendly stocks
  • I started to see that stocks have characters just like people. Some of them are calm, slow, conservative. Others are jumpy, nervous, tense.
  • And some of them I could not handle. Each time I bought them they did me an injury.
  • I began to take the view that if these stocks slapped me twice I would refuse to touch them any more. I would just shake off the blow and go away to buy something I could handle better.
• How to choose among short-listed stocks
  • Let the market decide which of the stocks are strong / weak.
  • Enter a small position in each of the stocks. Put a stop loss of 10% below buying price.
  • The stocks that get stopped out are discarded.
• Avoid fallen champions
  • A stock which was formerly a champion, selling at, say 150, may look like a bargain if it is currently on the market at 40 and gaining ground.
  • But I also figured that it is coming from behind, and thus laboring at a terrific handicap. To have slipped back from 150 to 40 means to have inevitably inflicted serious losses on all the traders who bought i at or near the peak and were later forced to sell at lower prices.
  • Thus there is certain to be strong psychological resistance to overcome, much ground to regain, before such a stock will again begin to look like a winner.
  • … I came to the conclusion that the only stocks which would be of real interest to me would be those that were breaking all previous records: stocks not merely rising in price, but actually in their highest boxes ever.

Darvas’ Principal Philosophy

• My only sound reason for buying a stock is that it is rising in price.
• If that is happening, no other reason is required.
• If that is not happening, no other reason is worth considering.

Darvas Box System Rules

Required Elements

• All-time high of the stock
• High and low for the past 2-3 years
• Weekly price range and volume for at least the last 4-6 months

Routine

1. Check general market trend as indicated by market averages
2. Examine 6-8 stocks of each of the 3-4 industries that he was interested in, to see how the industries behave in relation to the general trend of the market.
3. Examine the price changes in stocks he held and stocks he was interested in.
4. Review the stock page to see unusual price and volume changes for possible new candidates.

Short-listing Process (Techno-Fundamentalist approach)

• Price
  1. First stage filter – Short-list stocks where
     1. The high for the year is at least double the low for the year; AND
     2. The stock’s high for the week is at or within a few points of the high for the year.
  2. Second stage filter – Short-list stocks that were at their historical peaks.
• Volume
  • Stocks showing remarkably increased volume in a stock in which trading had been relatively quiet.
• Fundamentals
  • Capitalization: Volume of trading activity is large compared to the number of common shares outstanding [my note: comparing volume with shares float would be more accurate.]
  • Expected Earnings: Stocks with the greatest expectation of increasing earnings.
  • Industry Group: Stock is in an infant industry where earnings could double or treble, not in a static or dying industry.

Target Stocks

• A usually inactive stock which suddenly becomes active (volume should at least double), coupled with an advance in price.
• Stocks making new historical highs, not just highs over a lesser period (e.g. 5 years).
• Price keeps moving upwards with volume consistently high.
• It should be a lively stock, else it would probably not rise dynamically.
• It should move into a higher box. If it fell into a lower box, it is not behaving correctly and should be eliminated.

Establishing the Top and Bottom of Boxes

• The top of a box is established when the stock does not touch or penetrate a previously set new high for three consecutive days.
• The bottom of a box is established when the stock does not touch or penetrate a previously set new low for three consecutive days.
• The bottom of a box can only be established after the top of a box is established. The bottom can be established in the same day when the top was established.

Placing Orders

• On-stop purchase order is placed 1/8 above the top of the box, with a stop-loss order placed 1/8 below the top of the box.
• Keep the stop-loss order at its previous level until the stock had established the upper and lower level of its new box. Then raise the previous stop-loss order to 1/8 under the lower limit of the new box.
• Where the historic high is above a box high, the on-stop purchase order is set 1/8 above the historic high, and the stop-loss order is set 1/8 below the historic high.
• [My notes: When to Buy]
  • The above is the “official” rules from the books, however in page 71 of “How I Made $2,000,000 in the Stock Market”, Darvas did the following trades: Bought Baltimore & Ohio Railroad at 56.25 thinking that it was in the 56/61 box, bought Dobeckmun at 45 thinking that it was in a 44/49 box, bought Foster Wheeler at 61.75 thinking that it was in the 60/80 box.
  • Since Darvas does not predict stock prices, it is somewhat puzzling as to how he came up with the expected box tops of 61, 49, and 80 in the examples above.
  • One explanation could be that the box was formed and he bought it near the bottom of the box thinking that it would advance, hence the language use of “thinking that it was in the XX/YY box”.
• [My notes: Where to set Stop-loss]
  • While the book had the initial stop-loss placed 1/8 below the top of the box, in one of the Q&As on whether the stop-loss is too close, Darvas replied that the stop losses were placed either immediately following a massive breakthrough on the up side (in which case the stop-loss was placed just below the breakthrough point) or; a fraction below the bottom of a box.
  • It is not clear whether if a normal breakthrough (not a massive one) would work as well.
  • If you buy when the next box bottom (after breakout) is formed, you can have a good situation where the initial stop-loss is near.
  • This is assuming that the time distance between the breakout and the formation of the next box bottom is short. If the time distance is short, the chance that it will continue to break through the current box top is still good, so entry at the formation of the box bottom is fine.
  • However if the time distance is long, or the vertical price distance is long, the chance that it will break through the current box top can be significantly diminished. In this situation it is better to not take the risk, and wait for a situation where the distances are short.
  • One situation is where the breakout is particularly strong, e.g. stock shoots up by 15% due to some good news. If you want to buy on breakout, the initial stop loss should be set at the low made on the day the stock broke out, because the box top that was just broken would be too far away. In the situation where this particularly strong breakout coincides with the formation of the box bottom (after a previous breakout), the initial stop loss should also be set at the low made on the day the stock broke out, instead of the box bottom just made.

Modern Adaptation of Darvas Box by Daryl Guppy

Changes in Trading Situation from Darvas’ Time

• Higher volumes
• Higher volatility

Target Stocks

• Stocks making new high for the rolling 12- or 6-month period.

Basic Orders

• Buy order placed for the following day when the High or Close today exceeds the top of the Darvas box. This is because intra-day volatility tends to create false signals.
• Sell order placed for the following day when the Close today drops below the Darvas stop-loss point. While there is a chance that the stock can drop quickly, most likely you would be able to sell at the same Close price the following day. The benefit of this change is that it substantially reduces the number of false exit signals due to intra-day volatility.

Variation of Entry Rules

• Aggressive entry
  • After a box is formed, buy close to the bottom of the box, with a stop loss at the bottom of the box, in anticipation of a breakout.
  • This aims to capture more of the upside by buying early close to the support level (i.e. bottom of the box).
  • [My thoughts]
    • A precise rule wasn’t given, so I would think that buying immediately after the bottom of a box has been identified would be a good implementation of this rule because that is as “close” (time-wise) to the bottom of the box that you can get, and since you don’t predict the stock price movement, that point is as good as any others.
    • In addition, it would have been a point where the low is above the lowest low in the past 3 days, implying that it would have been a positive / neutral day. People tend to think that the market will go up the next day if the previous day was positive, so that would be a positive reason to using that as the buy point.
• Delayed entry
  • After there is a breakout of the top of a box, wait for a pullback that goes back into the box towards the bottom of the box.
  • Then buy close to the bottom of the box, with a stop loss at the bottom of the box, in anticipation of a reversal and move into a higher box.
  • [My thoughts]
    • This may be a good method to handle overlapping Darvas boxes, basically to buy when the bottom of the next Darvas box is formed. 
    • The benefit of this is that you solve the problem of being stopped out each time the next box’s bottom is below the box top you used to initiate the buy, which can result in accumulating losses even though the stock is trending up. 
    • The problem is solved because the stop loss is at the bottom of the box, hence so long as the bottom of the next box is higher, you don’t get stopped out. 
    • Essentially you have shifted the comparison to between the ‘bottoms’, versus comparing between the top (used to initiate the position) and bottom of the next box. 
    • The disadvantage of this is that stocks with overlapping Darvas boxes do mean that the strength of the up move is not that strong. So you are essentially playing with a weaker stock, and changing the system to fit this weaker stock.
    • For Darvas, note that only the first comparison is between the top and bottom, if a stock passes the initial test (i.e. showing that the stock is strong), thereafter the comparisons will be between bottoms.
    • Another potential way to deal with this is to increase the number of days from 3 to say 5 – 7. That will filter out more non-significant resistance and support, and does increase the vertical distance between boxes (i.e. less chance of overlapping boxes).
    • Of course, another potential is to do both, increase the window and buy at the bottoms.]

Variation for Trading Breakouts on Reversal (Breakout Darvas)

• After the breakout of a downtrend (i.e. reversal), initiate the trade after the formation and breakout of the 2nd Darvas box.

Handling Gap Ups of Darvas Boxes

• There is a problem when a stock breaks outs of a Darvas box, and moves up significantly without forming another Darvas box, because the stop loss point would still be at the bottom of the previous Darvas box, which would be a long way down.
• To protect profits in such a situation, imagine “Ghost Boxes” that stack on top of the previous Darvas box, with the same height as the previous Darvas box.
• Such Ghost Boxes are actualized when the stock price moves above the top of a Ghost Box. The stop loss is then raised to the bottom of these Ghost Boxes.
• Hence when the first Ghost Box is created (after the stock price exceeds the top of this Ghost Box), the stop loss will essentially be raised to the top of the previous Darvas box.
• Subsequent Ghost Boxes being created will also mean that the stop loss will get raised to the top of the previous Ghost Box.

-END-