Darvas Box Strategy: How Nicolas Darvas Traded Breakouts

Who was Nicolas Darvas?

Nicolas Darvas (1920–1977) was a Hungarian-born dancer who trained as an economist in Budapest, left Hungary in 1943 and later toured the world as half of a ballroom dance team. He taught himself the market from books and, by his own account, never sold short: he only bought.

Touring meant no live quotes. He followed his stocks through printed price tables in financial newspapers and sent buy and sell stop orders to his New York broker by cable and telegram. Those limits shaped the method: rules simple enough to set in advance and leave working while he was on stage.

He told the story in his 1960 book, How I Made $2,000,000 in the Stock Market. It is a personal account, not verified results. Read it for the method, not the numbers. A famous trader’s results are not typical, and the method fails often.

The book follows his stages as an investor: the gambler, the fundamentalist, the technician and, last, the techno-fundamentalist, the name of its final part, where he combines boxes with a look at the business.

What is a Darvas box? How to draw the top and bottom

Darvas saw a rising stock as a stack of boxes. After a strong advance, price stalls and swings inside a range; that range is the box. While price stays inside, there is nothing to do. A move out of the top suggests the advance is resuming; a move out of the bottom says it has failed.

Hypothetical: a stock in an uptrend sets a high of $48.00 on a Monday, and the next three sessions all trade below it. It then dips to $44.50 and holds above that low for three more sessions. Box 1 runs from $44.50 to $48.00, about 7% deep. The drawing at the top of this page shows three such boxes stacked like stairs.

Darvas drew his boxes by eye from daily highs and lows. Later writers turned his description into a mechanical rule, usually summarized like this:

1. The stock makes a new high, in modern versions often a new 52-week high.
2. If the next three sessions fail to trade above that high, it becomes the box top.
3. Price pulls back. The lowest low that then holds for three sessions becomes the box bottom.
4. The box stays valid until price leaves it. A break above the top starts the next box higher; a break below the bottom ends the trade.

Box top

The high that price failed to exceed for three sessions. The buy level sits just above it.

Box bottom

The lowest low that held for three sessions after the top formed. The trailing stop goes just under it.

Breakout

A move out of the box. Darvas acted only on breaks above the top; a break below the bottom was his signal to be out.

Stacked boxes

In a healthy uptrend each new box forms higher than the last. A new box that overlaps or sits lower says the trend is tiring.

Darvas’s techno-fundamental filter: growing industries and earnings

In his techno-fundamental stage, the boxes and the volume decided when to buy, and a look at the business narrowed what to buy. He favoured companies in young, fast-growing industries where new products could lift earnings; electronics and rocketry were the themes of his day.

His fundamental check was a filter, not a valuation model: is this a company tied to the future, and should its earnings rise? The drawing below shows the logic: revenue and earnings per share (EPS) climbing over five years, with the share price following earnings over time but not every year.

A modern version, hypothetical: Company A grew revenue 25% and EPS 30% over the past year, and its last two quarters grew faster than the two before. Its chart sets a new high on heavy volume. Company B is breaking out too, but its earnings are shrinking. Darvas’s approach favours A. Rising earnings are not a buy signal by themselves; they only decide which breakouts deserve your attention. CAN SLIM is a later, more detailed version of this blend.

The Darvas buy rule: a close above the box top on volume

Darvas entered with an automatic buy-stop order placed just above the box top, so he was filled as soon as the breakout traded. He wanted the move to come on volume clearly above normal, a sign that buyers were arriving in size.

With today’s charts, many traders wait for the daily candle to close above the top instead. A close filters out intraday pokes that reverse by the end of the day, at the cost of a slightly worse price. Volume is easy to check with relative volume (RVOL): the day’s volume divided by its average volume, for example over 50 days.

Hypothetical, continuing box 1: the stock closes at $48.60, above the $48.00 top, on 3.6 million shares against a 1.2 million average: RVOL 3.0, like the tall bar in the drawing below. That is the breakout Darvas looked for. A close above the top on RVOL 0.8 is weaker; many traders skip it or wait to see whether price holds above the old top.

Location matters too. A breakout from a tight box near a new high has little overhead supply. A breakout that has already run 15% past the box before you notice it leaves the stop far away. The breakout trading strategy guide covers false-break filters in more detail.

Where the stop goes: under the box, then trailed up

Darvas placed his stop just under his purchase price right after buying, so a failed breakout cost little. As the stock rose and formed a new box, he raised the stop to just under the new box’s bottom. The stop only moved up, never down. Adding shares as the trend extends is a separate decision, covered in Jesse Livermore’s trading rules.

Many modern versions start the stop under the bottom of the box that was just broken: it survives more noise but risks more per share. Neither is right for everyone; what matters is choosing before the trade and sizing from it. Hypothetical, from the $48.60 entry:

• Tight, Darvas-style: $47.40, just under the old $48.00 top. Risk $1.20 per share.
• Wide: $44.40, just under the $44.50 box bottom. Risk $4.20 per share.
• Box 2 forms at $51.80–$55.00: either stop moves up to $51.70. A stop-out now would close the trade $3.10 above the entry, before costs and gaps.
• Box 3 forms at $55.30–$59.50 after a close above $55.00: the stop moves to $55.20.

The retest and the old box top

After a clean breakout, the old box top often acts as a floor: price dips back to it, holds and turns up, as in the drawing below. That retest is an alternative entry for traders who missed the breakout or want a tighter stop, placed just under the retest low. It does not always come; strong breakouts often never look back.

If price instead closes back inside the box, the breakout has failed, and the tight stop is usually already gone. The crossed-out short in the drawing is the other lesson: once price closes outside a range, the range is over, and selling its old top is fighting the new trend. For trading inside a box rather than out of it, see the range trading strategy.

Position size: risk a fixed percentage on every box

The stop sets the risk per share; your account sets the risk per trade. Shares = dollar risk ÷ (entry − stop). With a fixed percentage, every stop-out costs about the same, whether the stop is tight or wide.

Hypothetical: a $25,000 account risking 1% per trade accepts a $250 loss. With the tight stop ($1.20 risk), $250 ÷ $1.20 = 208 shares, a $10,109 position. With the wide stop ($4.20 risk), $250 ÷ $4.20 = 59 shares, a $2,867 position. Same dollar risk, very different exposure.

Now follow the trade to the exit. The stock tops out inside box 3 and trades down through $55.20, closing the trade about $6.60 a share above the entry: roughly +5.5R for the tight-stop version and +1.6R for the wide one. On another chart, the first dip hits the tight stop and the wide one survives. You cannot know which in advance, so the size, not the stop, keeps each loss at 1R. The position size calculator does the math.

• Choose the risk per trade first; many traders keep it between 0.5% and 1% of the account.
• Measure entry to stop, divide, and round the share count down.
• Check the position value: a very tight stop can produce a position larger than you can hold.
• Allow for gaps: a stop order becomes a market order once triggered and can fill below your price.

When Darvas boxes fail: false breakouts, whipsaws and chop

Darvas boxes fail often, and on the day a failure looks exactly like a good breakout. Same hypothetical stock, another path: the $48.60 close on RVOL 3.0 is followed by a close at $47.20, back inside the box. The tight stop at $47.40 is hit: 208 × $1.20 = $249.60 lost, −1R, as planned.

The bigger danger is a choppy market. In a sideways tape, stocks poke above box tops and fall back again and again. Three failed breakouts in a row at 1% each is −3R, or $750 on the $25,000 account: survivable because the risk was fixed, ruinous if each loss had been allowed to grow.

Copying Darvas’s rules does not copy his outcome; markets, costs and stocks have changed since the 1950s. Treat the boxes as a way to define risk, not a forecast. The usual ways they go wrong:

• False breakouts: a close above the top that reverses within a day or two. Waiting for a close and for high RVOL helps but does not prevent them.
• Whipsaws in chop: when the broad market goes sideways or down, most breakouts fail. Many breakout traders cut size or stand aside until the index trends again.
• Gaps: an earnings gap can open far below your stop, and the loss can then exceed 1R.
• Late entries: buying a breakout that has already run far above the box puts the stop far away. Skip it or wait for the next box.
• Subjective boxes: two traders can draw different tops and bottoms on the same chart. Write the three-day rule into your plan and apply it the same way every time.
• Hindsight: old charts with perfect staircases are picked after the fact. Many boxes break down instead of stacking up.

How to apply the Darvas box strategy with today’s tools

Darvas worked from newspapers and telegrams; you have screeners, charts and alerts. The method still comes down to four questions: is this a strong stock in a rising market, where is the box, did price close above the top on volume, and where is the stop?

It is a daily-chart, swing-to-position method: trades can last weeks or months. The same idea, a range, a close outside it and a stop on the other side, also appears on intraday charts, where boxes form and break much faster and false breakouts are more frequent.

If you day trade that idea on 5-minute to 1-hour charts, UpTickFlow, our invite-only TradingView indicator, works on a similar close-first principle, so weigh this accordingly. It does not draw Darvas boxes or screen stocks. It waits for a candle to close at a level, draws an entry, a stop and a target before the trade, scores the setup from 0 to 100 and stays quiet in chop: a second check, not a scanner. For the daily method, the steps:

1. Build a watchlist of stocks at or near 52-week highs in an uptrend. A common modern filter is drawn below: price above rising 50-, 150- and 200-day moving averages, stacked in that order, and close to the 52-week high.
2. Add a fundamental check in the Darvas spirit: rising revenue and earnings in a growing industry.
3. Draw each box on the daily chart with TradingView’s Rectangle tool once its top and bottom are confirmed.
4. Set an alert just above the box top so you don’t have to watch all day; the TradingView alerts guide shows how.
5. On a close above the top with RVOL well above 1, enter, with the stop and the size at your fixed percentage decided before the order.
6. Raise the stop under each new box bottom, never lower it, and log every trade in R.