// STUDY RESULT · 1998-01 TO 2026-09

Inspired by Bollinger: Band Squeeze and Breakout

NO EDGE

Bands placed around a moving average scale price dispersion by recent volatility. A contraction in band width can flag a low-volatility regime before a directional move, but the bands alone do not determine direction.

How this run differs from the source: ATR contraction stands in for a BandWidth low and a 20-day high for the upper band; in cash two months in three.

	THE STRATEGY	$SPXTR BUY & HOLD
CAGR	0.0%	9.3%
Sharpe ratio	0.07	0.66
Max drawdown	-45.2%	-50.9%

// GROWTH OF $1, LOG SCALE

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

Out-of-sample (from the split): CAGR -0.3% against 14.0% for the benchmark; Sharpe 0.09 in-sample, 0.00 out-of-sample.

After 25 bps of trading costs: -3.2% a year. EDGE OVER RANDOM

// DECADE BY DECADE

ERA	STRATEGY	BENCHMARK
1990s	12.6%	20.9%
2000s	-3.1%	-0.8%
2010s	0.0%	14.0%
2020s	1.2%	15.2%

// YEAR BY YEAR, LOSING YEARS IN RED

1998*15.9%27.2%

19999.5%21.0%

200010.1%-9.1%

2001-10.8%-11.9%

2002-17.6%-22.1%

20036.0%28.7%

2004-6.8%10.9%

20051.2%4.9%

200612.6%15.8%

2007-1.1%5.5%

2008-13.8%-37.0%

2009-2.0%26.5%

20100.6%15.1%

2011-4.6%2.1%

201211.1%16.0%

20132.9%32.4%

2014-2.0%13.7%

2015-1.4%1.4%

2016-4.9%12.0%

20170.0%21.8%

20184.9%-4.4%

2019-7.5%31.5%

202016.9%18.4%

2021-8.1%28.7%

20227.6%-18.1%

2023-6.1%26.3%

2024-2.0%25.0%

20252.8%17.9%

2026*-2.6%11.8%

Losing years: 2001 (-10.8%), 2002 (-17.6%), 2004 (-6.8%), 2007 (-1.1%), 2008 (-13.8%), 2009 (-2.0%), 2011 (-4.6%), 2014 (-2.0%), 2015 (-1.4%), 2016 (-4.9%), 2019 (-7.5%), 2021 (-8.1%), 2023 (-6.1%), 2024 (-2.0%), 2026 (-2.6%). Each cell: the strategy, then the benchmark below it. * a part year.

// THE IDEA, AS PUBLISHED

John Bollinger (2001), Bollinger on Bollinger Bands · 2001

20-period SMA of closes with bands 2 standard deviations above and below
Squeeze: BandWidth ((upper - lower) / middle) at its lowest in about 6 months
Enter on a close outside a band after the squeeze, confirmed by a volume indicator; beware a first false move
Exit on a trailing stop such as Parabolic SAR or a tag of the opposite band

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 08:46 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