// STUDY RESULT · 1998-01 TO 2026-09

Inspired by Darvas: The Darvas Box

TRADE-OFF

Stocks move in stacked 'boxes'; buy the break of the box ceiling at new highs with a stop just below it.

How this run differs from the source: No box is drawn: a 20-day volume breakout near the 52-week high stands in, with no earnings test or stop.

	THE STRATEGY	$SPXTR BUY & HOLD
CAGR	12.5%	9.3%
Sharpe ratio	0.52	0.66
Max drawdown	-74.8%	-50.9%

// GROWTH OF $1, LOG SCALE

AMBER: THE STRATEGY · GREY: $SPXTR BUY & HOLD

Out-of-sample (from the split): CAGR 9.3% against 14.0% for the benchmark; Sharpe 0.56 in-sample, 0.46 out-of-sample.

After 25 bps of trading costs: 7.0% a year. BELOW RANDOM

// DECADE BY DECADE

ERA	STRATEGY	BENCHMARK
1990s	127.8%	20.9%
2000s	0.3%	-0.8%
2010s	3.5%	14.0%
2020s	22.3%	15.2%

// YEAR BY YEAR, LOSING YEARS IN RED

1998*47.9%27.2%

1999236.1%21.0%

2000-41.6%-9.1%

2001-9.4%-11.9%

2002-24.3%-22.1%

2003112.8%28.7%

20048.2%10.9%

20059.4%4.9%

20069.1%15.8%

200724.3%5.5%

2008-27.3%-37.0%

200916.2%26.5%

20106.0%15.1%

2011-17.2%2.1%

201237.9%16.0%

201355.6%32.4%

2014-1.2%13.7%

2015-16.2%1.4%

2016-8.0%12.0%

20175.4%21.8%

2018-20.7%-4.4%

201914.2%31.5%

202049.9%18.4%

202125.3%28.7%

2022-14.4%-18.1%

20232.0%26.3%

202454.2%25.0%

202580.4%17.9%

2026*-18.2%11.8%

Losing years: 2000 (-41.6%), 2001 (-9.4%), 2002 (-24.3%), 2008 (-27.3%), 2011 (-17.2%), 2014 (-1.2%), 2015 (-16.2%), 2016 (-8.0%), 2018 (-20.7%), 2022 (-14.4%), 2026 (-18.2%). Each cell: the strategy, then the benchmark below it. * a part year.

// THE IDEA, AS PUBLISHED

Nicolas Darvas, 'How I Made $2,000,000 in the Stock Market' (1960) · 1960

Stocks at new highs on unusual volume, in growth industries with rising earnings
Box top: a high not exceeded for about 3 days; floor: a low that then holds for about 3 days
Buy stop just above the box top as price enters a new box
Stop just below the latest box floor, raised as boxes stack

RUN IT YOURSELF ▸ IN THE LIBRARY

A real computation on the QuantGPT warehouse, walk-forward. A study across the whole market includes only part of the companies that later delisted from 1998 (survivors only before); a study on your own list of tickers is not affected. Gross returns, no costs except where stated. Past performance is not a promise. Findings, not advice. COMPUTED 2026-11-07 08:41 ON THE QUANTGPT WAREHOUSE. COMPANIES THAT LATER DELISTED ARE ONLY PARTLY IN THIS UNIVERSE (A FIX IS IN PROGRESS), SO THE RESULT LEANS TOWARD SURVIVORS. FUNDAMENTALS BY FILING DATE. NOT INVESTMENT ADVICE. PRIVACY