// STUDY RESULT · 1998-01 TO 2026-09

Inspired by George and Hwang: 52-Week High Momentum

NO EDGE

Nearness to the 52-week high forecasts returns better than past returns, and its gains do not reverse later; the authors attribute this to investors anchoring on the 52-week high when judging news.

How this run differs from the source: Many names tie at a new 52-week high, so the 20 held are partly picked in ticker order; the paper holds the top 30% long-short.

	THE STRATEGY	$SPXTR BUY & HOLD
CAGR	-0.3%	9.3%
Sharpe ratio	0.05	0.66
Max drawdown	-55.9%	-50.9%

// GROWTH OF $1, LOG SCALE

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

Out-of-sample (from the split): CAGR -0.2% against 14.0% for the benchmark; Sharpe 0.05 in-sample, 0.04 out-of-sample.

After 25 bps of trading costs: -5.6% a year. BELOW RANDOM

// DECADE BY DECADE

ERA	STRATEGY	BENCHMARK
1990s	4.0%	20.9%
2000s	-3.2%	-0.8%
2010s	-0.1%	14.0%
2020s	2.8%	15.2%

// YEAR BY YEAR, LOSING YEARS IN RED

1998*11.0%27.2%

1999-2.1%21.0%

20005.1%-9.1%

2001-0.7%-11.9%

2002-10.3%-22.1%

200328.3%28.7%

2004-8.8%10.9%

2005-8.1%4.9%

2006-5.3%15.8%

2007-0.1%5.5%

2008-33.1%-37.0%

200913.8%26.5%

20106.6%15.1%

2011-11.9%2.1%

20126.1%16.0%

201318.1%32.4%

20144.2%13.7%

2015-9.0%1.4%

2016-9.8%12.0%

20173.4%21.8%

2018-11.8%-4.4%

20196.0%31.5%

202040.4%18.4%

2021-5.4%28.7%

2022-12.1%-18.1%

2023-10.7%26.3%

20243.4%25.0%

202514.3%17.9%

2026*-1.4%11.8%

Losing years: 1999 (-2.1%), 2001 (-0.7%), 2002 (-10.3%), 2004 (-8.8%), 2005 (-8.1%), 2006 (-5.3%), 2007 (-0.1%), 2008 (-33.1%), 2011 (-11.9%), 2015 (-9.0%), 2016 (-9.8%), 2018 (-11.8%), 2021 (-5.4%), 2022 (-12.1%), 2023 (-10.7%), 2026 (-1.4%). Each cell: the strategy, then the benchmark below it. * a part year.

// THE IDEA, AS PUBLISHED

George & Hwang (2004), Journal of Finance · 2004

Rank stocks by current price divided by the highest price of the past 12 months
Buy the top 30 percent (nearest the high), sell the bottom 30 percent, equal-weighted
Hold 6 months in overlapping monthly cohorts

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 08:42 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