Darvas Box Theory: How the Trading Strategy Works

Search Fintrev

Search our guides, reviews and glossary.

The Darvas box theory is a momentum trading strategy: buy a stock only after its price breaks above a well-defined trading range on strong volume, then protect the position with a stop-loss that rises as the trend continues. Nicolas Darvas developed the method in the 1950s. He drew a “box” around a stock’s recent highs and lows, treated the top as the breakout level and the bottom as the floor, and moved his stop higher each time the price climbed into a new box. The system does not predict where a stock will go. It follows a trend already underway and exits when that trend fails.

Who Nicolas Darvas was

Darvas was born in Hungary in 1920 and studied economics, but his career was dance. After reaching the United States in the 1940s, he toured nightclubs and theaters worldwide with his half-sister Julia as “Darvas and Julia.”

His first stock purchase came in 1952: an impulsive buy of a Canadian company that made a small profit and started a long fascination with markets. Touring through most of the 1950s, he could not watch prices in real time. He followed the market through weekly copies of Barron’s and The Wall Street Journal and their printed price and volume listings, which pushed him toward broader price behavior rather than intraday noise.

Between 1957 and 1959 he reportedly grew $10,000 into more than $2 million. Time featured him in 1959, after the S&P 500 had risen more than 50% in two years. In 1960 he published How I Made $2,000,000 in the Stock Market, a bestseller that made trading seem open to ordinary people. The book remains the main source for his rules.

How the Darvas box system works

Choosing which stocks to watch

Darvas did not apply boxes to every stock. He looked for companies in industries he expected to excite investors, sectors tied to new technology and future growth that could “climb into the stratosphere.” The fundamental indicator he looked for above all was improving earning power, or the anticipation of it. In choppy markets he leaned toward companies with solid earnings records.

His reasoning had little to do with intrinsic value: stocks tied to future development drew investor enthusiasm, and his job was to follow those fashions and get out before they changed. He kept a short watchlist and tracked each stock’s daily prices and volume.

Using volume as the alert

For Darvas, an unusual jump in trading volume was the main sign that a stock might be ready to move. Heavy volume alongside a new high suggested real buying interest. A new high on thin volume did not carry the same weight. Volume was not a separate indicator in his system. It was the condition that made a price signal worth acting on.

Drawing the box

A box marks a consolidation: a period when the price pauses inside a narrow range after making a new high.

1. Find the ceiling. The stock makes a new high. If the price then fails to exceed that high for several sessions, the high becomes the top of the box. One standard formalization of his rules uses three consecutive sessions that fail to go higher.
2. Find the floor. After the ceiling is in place, identify the lowest low of the pullback. When that low holds for the same confirmation period, it becomes the bottom of the box.
3. Wait. While the price trades between the floor and the ceiling, nothing happens. The box is just a range.

If the price breaks above the ceiling before the floor is confirmed, the box is not finished and the process restarts from the new high. A Darvas box indicator drawn on live data can therefore change shape until both edges are confirmed, so a box that looks clean in hindsight may not have existed in that form in real time.

The ceiling and floor resemble support and resistance levels, except that Darvas cared only about ranges in an uptrend and used them as entry and exit triggers rather than zones to trade back and forth.

Entering, trailing and adding

When the price breaks above the box ceiling, Darvas bought. He then used the ceiling of the box that had just been broken as the reference for his stop-loss. As the stock formed a new, higher box and broke out of that one too, he raised the stop again and often added to the position. The trade usually ended only when the stop was hit.

He summed up the approach in his book: he would jog along with an upward trend, trailing his stop-loss behind him, buy more as the trend continued, and “run like a thief” when it reversed. The rule set is asymmetric by design. Losing trades are cut quickly near the breakout level, while winning trades stay open as long as the boxes keep stepping higher.

Traders have since adapted the stop rule. Some place the stop just under the broken ceiling, as Darvas described. Others place it below the floor of the most recent box, which gives the trade more room but risks more per share. The choice changes position size considerably, as the examples below show.

Darvas box examples

All three examples are hypothetical and exist only to show the mechanics. Figures ignore commissions, spreads and slippage, and assume a trader willing to risk $500 per trade.

Example 1: the first breakout

A stock makes a new high of $48.00. Over the next three sessions its highs are $47.60, $47.90 and $47.20, so $48.00 becomes the ceiling. The pullback low is $45.50, and the following three sessions all hold above it, so $45.50 becomes the floor. The box runs from $45.50 to $48.00.

A few days later the stock trades up to $48.20 on volume roughly twice its recent average. The trader buys at $48.20.

• Stop just under the ceiling at $47.80: risk is $0.40 per share. A $500 risk budget allows 1,250 shares ($500 ÷ $0.40).
• Stop just under the floor at $45.40: risk is $2.80 per share. The same budget allows about 178 shares ($500 ÷ $2.80).

The tighter stop allows a much larger position but will be triggered by ordinary pullbacks more often. The wider stop survives more noise but earns less on the same move. A position size calculator handles this arithmetic once entry, stop and risk amount are set.

Example 2: trailing the stop through a second box

The stock from Example 1 climbs to $53.00, stalls, and forms a new box from $50.10 to $53.00. The trader raises the stop toward the old ceiling area or the new floor, depending on their rule, and no longer risks the original entry. When the price breaks above $53.00 on heavy volume, the trader may add shares, sizing the addition so the total position still fits the risk budget at the new stop. Each successful box moves the worst-case exit higher.

Example 3: a failed breakout and a gap

A different stock breaks above a $48.00 ceiling to $48.30, but volume is no higher than normal. Two sessions later it falls back to $47.50, and the stop at $47.80 is triggered. This is the small loss the system accepts as the cost of being in position for the breakouts that do run.

Now suppose instead that bad news hits overnight and the stock opens at $47.00. A standard stop order at $47.80 becomes a market order once triggered and fills near the opening price, not at $47.80. The actual loss per share is about $1.20, three times the planned $0.40. A stop-limit order would avoid the low fill but might not fill at all, leaving the position open as the price keeps falling. The box defines the trigger. It does not guarantee the fill price.

The “techno-fundamentalist” side of the method

The Darvas box is usually described as pure technical analysis, but Darvas used fundamentals to decide which stocks deserved attention, and price and volume to decide when to act. His objection to fundamental analysis was mainly about time. In his book he tells of recommending a stock to his secretary, whose father insisted on studying the company’s books first. While the father was still reviewing them, the stock climbed to $50, a 40% gain the family missed. A modern version follows the same two steps: a growth or earnings filter first, then a box breakout on rising volume.

Discipline and record keeping

Darvas stressed that the rules only work if they are followed without exception. He logged every trade and analyzed what went right and wrong, which is how he refined the system. The box method’s value is its fixed structure: entry, initial stop and stop adjustments are all defined in advance by the chart, leaving less room for judgment calls made under pressure. Practicing in a stock market simulator lets you draw boxes, place stops and keep a trade log without financial risk.

Applying Darvas boxes to forex and other markets

The box structure works on any chart, including currency pairs, indexes and crypto, but two adjustments matter in forex.

First, spot forex has no centralized exchange, so no single, market-wide volume figure exists. Platforms typically show tick volume, the number of price changes in a period, which is the broker’s own feed rather than total traded volume. Traders using a Darvas box strategy on currency pairs often treat tick volume as a rough proxy or rely on other confirmation, such as the trading session in which the breakout occurs. Breakouts during thin hours behave differently from those during the overlap of major forex trading sessions.

Second, the time frame changes the trade. Darvas worked from daily data and held winners for as long as new boxes kept forming. On daily or four-hour charts, the method resembles swing trading, with positions held for days or weeks. On intraday charts it becomes a short-term breakout method closer to day trading, with many more boxes and false breakouts, and costs take a larger share of each move.

Limitations of the Darvas box theory

• It was built in a bull market. A long-only breakout system does best when many stocks are trending up, as they were from 1957 to 1959.
• Failed breakouts add up. When trends do not develop, the method produces a sequence of small losses. Several in a row can erase much of a single good trade, and they test the trader’s willingness to keep following the rules.
• It lags by design. Each box needs confirmation before it exists, so entries come after the move has started, and exits come after the trend has already turned.
• Stops are not guaranteed. As Example 3 shows, gaps and fast markets can fill a stop well below its level.
• Box rules vary. The confirmation period, the stop placement and what counts as “heavy” volume differ across books, indicators and traders. Two people looking at the same chart can draw different boxes, and backtest results depend on the exact rules used.
• The original record is self-reported. The $10,000 to $2 million figure comes from Darvas’s own account and press coverage at the time.

Similar trading approaches

Several methods share the Darvas box’s core logic of trading breakouts from defined ranges:

• Turtle trading. Developed in the 1980s, this system enters when price exceeds the high (or falls below the low) of a set number of past periods, a mechanical breakout rule similar to a box ceiling, applied in both directions.
• Wyckoff method. Richard Wyckoff also studied trading ranges and volume, but read them as phases of accumulation and distribution by large buyers and sellers. Darvas simply waits for the range to break.
• Classic trend following. Many trend systems buy when price is above a moving average and sell when it drops below. The moving average plays the role that the trailing box plays for Darvas.

FAQs

The Bottom Line

The Darvas box suits traders who want fixed, chart-based rules for entering breakouts and trailing a stop behind a rising trend. It works best when many stocks are trending up and tends to produce strings of small losses in sideways or falling markets, so a fundamental filter, volume confirmation and a complete trade log all matter. The box sets the exit trigger, yet gaps can still fill a stop well below its planned level.