52-Week High Proximity: One Signal, 18 Markets, Very Different

India returned 26.73% annually. The UK returned 2.93%. Same signal. Same parameters. Same rebalance dates. The gap isn't explained by luck or sample size. It reflects a fundamental difference in who trades these markets and how they process information.

Contents

1. Method
2. Full Results
3. Where It Works: Asia-Pacific Retail Markets
4. India (BSE+NSE): 26.73% CAGR, +18.72% excess
5. Thailand (SET): 12.55% CAGR, +4.54% excess, 46.8% down capture
6. Korea (KSC): 11.43% CAGR, +3.42% excess, 39.1% down capture
7. Japan (JPX): 8.89% CAGR, +0.88% excess, 63.5% down capture
8. Where It Doesn't Work: European Institutional Markets
9. Germany (XETRA): 6.58% CAGR, -1.44% excess
10. UK (LSE): 2.93% CAGR, -5.08% excess
11. Switzerland (SWX): 5.29% CAGR, -2.72% excess
12. Italy (BIT): 6.89% CAGR, -1.12% excess
13. The Sweden Anomaly
14. The China Data Point
15. The Retail vs Institutional Divide
16. Down Capture: The Clearest Pattern
17. What This Means for Portfolio Construction
18. Part of a Series
19. References

The 52-week high proximity strategy works where anchoring matters. Anchoring is a retail bias. Individual investors treat the 52-week high as a psychological ceiling, hesitate to buy through it, and create systematic underreaction that resolves upward. That behavioral pattern is strongest in markets where retail investors dominate trading. It is weakest in markets where algorithmic trading, institutional funds, and professional managers set prices.

The geographic split in our 25-year backtest tracks almost perfectly with market structure. Asia-Pacific retail-heavy markets: strong results across the board. European institutional markets: little to no effect. The signal is real. Where it fires depends on who you're trading against.

Method

Same screen across all 18 exchanges. The only variables are the stock universe and the local market cap threshold.

For methodology detail, SQL, and the US year-by-year breakdown, see the US post.

Full Results

*Sweden: see caveat below.

Where It Works: Asia-Pacific Retail Markets

India (BSE+NSE): 26.73% CAGR, +18.72% excess

The standout result, by a wide margin. India's retail investor base is large and growing, and the 52-week high functions as a powerful psychological anchor in the market. The BSE Sensex era from 2002 to 2008 was extraordinary: the index went from approximately 2,900 to 21,000 points, and the proximity screen captured that rally with high efficiency.

Three years deserve direct mention: 2003 (+119.1%), 2005 (+84.4%), and 2007 (+112.2%). These aren't artifacts. They reflect the real BSE bull market of that period, when India's liberalizing economy attracted capital at a pace the market had never seen. The proximity signal fired consistently throughout, identifying the stocks leading each leg of the advance.

The down capture of 23.4% is remarkable. In bear markets, the strategy absorbed less than a quarter of the benchmark's losses. India also shows 10% cash periods, concentrated in 2000-2001 before the market developed enough qualifying stocks.

The MaxDD of -54.1% is the honest caveat. Drawdowns in India are deep when they hit. Volatility is high. The Sharpe of 0.735 is strong given this volatility, but investors need to be comfortable with that kind of peak-to-trough experience.

Thailand (SET): 12.55% CAGR, +4.54% excess, 46.8% down capture

Thailand's retail-heavy market produced solid results. The down capture of 46.8% shows the same defensive characteristic as India and Korea, just less pronounced. Cash periods accounted for 14% of the history, concentrated in 2000-2002 during the aftermath of the 1997 Asian financial crisis.

Korea (KSC): 11.43% CAGR, +3.42% excess, 39.1% down capture

Korea produced our second-lowest down capture of any market at 39.1%. Retail participation in the Korean equity market is among the highest in the world, and the psychological anchoring effect is clear in the data. Win rate of 50.5% is modest, but the asymmetry between winning and losing periods (down capture below 40%) explains the CAGR advantage.

Korea had cash periods accounting for 19% of the history, primarily in 2000-2004 when the post-dot-com environment didn't generate enough qualifying stocks.

Japan (JPX): 8.89% CAGR, +0.88% excess, 63.5% down capture

Japan barely outperformed. The Sharpe (0.548) is actually strong for such modest excess returns. The MaxDD of -43.0% is real, driven by Japan's structural deflationary period in the early 2000s. The down capture of 63.5% is similar to the US result.

Japan sits between the two clusters. It has meaningful retail participation, particularly in individual stocks, but also a substantial institutional base and algorithmic trading infrastructure. The result is consistent with that intermediate market structure: the signal fires, but weakly.

Where It Doesn't Work: European Institutional Markets

Germany (XETRA): 6.58% CAGR, -1.44% excess

German equity markets are heavily institutional. The retail participation rate is among the lowest in the developed world. The 52-week high as a psychological anchor has less grip when the marginal buyer is a quant fund or a pension manager running systematic strategies. Those buyers don't hesitate at a new high because of anchoring bias. They buy or sell based on model signals.

The result: -1.44% excess over 25 years. Not catastrophic, but the signal clearly has no edge here.

UK (LSE): 2.93% CAGR, -5.08% excess

The worst result in the dataset. The LSE is the most institutionally dominated large exchange we tested. Major UK equities are heavily covered, efficiently priced, and traded by professionals. The proximity signal finds nothing to exploit. 2.93% annual return is well below SPY's 8.01%, and the gap compounds badly over 25 years.

Switzerland (SWX): 5.29% CAGR, -2.72% excess

Similar story to Germany and the UK. Switzerland's market is small, concentrated in large global companies (Nestle, Novartis, Roche, UBS), and heavily institutionally owned. No anchoring effect visible.

Italy (BIT): 6.89% CAGR, -1.12% excess

Italy has higher retail participation than Germany or Switzerland but still underperforms. The Italian market's structural issues, concentrated in financials and industrials, and its political risk premium, likely overwhelm any behavioral edge.

The Sweden Anomaly

Sweden (STO) requires a direct caveat.

The nominal result looks positive: 8.67% CAGR, technically above SPY's 8.01%. But the down capture ratio is 131.3%. That means the strategy amplified losses in every major bear market.

2002: strategy -30.6% vs SPY -19.9%. 2008: strategy -38.0% vs SPY -34.3%. 2022: strategy -32.0% vs SPY -19.0%.

The positive nominal CAGR is entirely driven by exceptional gains in strong bull markets, which then get almost fully reversed in the next downturn. This isn't a strategy characteristic we'd recommend. The distribution of returns is adversarial for actual investors: big drawdowns happen at the worst times (when you least want them), and they're worse than just holding the market.

The likely explanation is structural. Sweden's equity market is heavily concentrated in a small number of growth and technology names. Many of the same names appear near their 52-week highs during bull runs, creating a concentrated portfolio that amplifies both upside and downside. We report Sweden for completeness, but treat it as a warning.

The China Data Point

China (SHH+SHZ) returned 8.30%, barely above SPY at +0.29% excess.

The 2007 data deserves direct mention: the strategy returned approximately 217.3% that year on Chinese exchanges. This is a real number. China's A-share market experienced a speculative bubble in 2006-2007 that drove the Shanghai Composite from roughly 1,200 to 6,100. Stocks near their 52-week highs in that environment were stocks still leading the bubble. The proximity signal happened to be in the right stocks through that entire run.

The bubble collapsed in 2008, and the losses were severe. The net effect over the full period is a modest excess return. China's market structure is unusual: substantial retail participation (which should help the signal) but also significant state influence and information asymmetry (which distort price signals). The mixed result reflects that complexity.

The Retail vs Institutional Divide

The pattern across all 18 markets is consistent enough to be structural, not coincidental.

Markets where individual investors account for a high share of trading volume (India, Korea, Thailand, and to a lesser extent the US and Japan) show the strongest results. The 52-week high functions as a real anchor in these markets because retail investors actually anchor to it.

Markets where institutional investors dominate (UK, Germany, Switzerland) show no edge or negative excess returns. Professional money managers don't anchor to the 52-week high as a psychological barrier. They may use it as a signal input, but they don't experience the hesitation that creates the behavioral gap the strategy exploits.

This has practical implications. The strategy isn't a universal momentum screen. It's a specific exploitation of a retail behavioral bias. Its effectiveness depends entirely on that bias being present in the market you're trading. Running this screen on the London Stock Exchange is trading against counterparties who aren't subject to the bias. Running it on the BSE is a different proposition.

Down Capture: The Clearest Pattern

The most consistent finding across all markets where the strategy worked is the low down capture ratio.

The mechanism is the same everywhere the signal works. Bear markets push stocks away from their 52-week highs. The proximity ratios fall. The qualifying universe shrinks toward the cash threshold. The portfolio naturally de-risks without requiring any explicit timing call. This is the behavioral finance mechanism operating in reverse: the same anchoring that creates the buying hesitation on the way up creates a natural exit signal on the way down.

What This Means for Portfolio Construction

Use this signal in markets where retail investors set prices at the margin. India and Korea are the clearest cases. Thailand is a reasonable third choice. The US produces solid results with a large, stable universe.

Avoid this signal in European institutional markets. The LSE, XETRA, and SWX results aren't bad luck. They're consistent with the theory. There's no behavioral gap to exploit when professionals are the marginal buyer.

Treat the Japan result as interesting but modest. Japan's hybrid market structure produces a weak version of the effect. Usable, but not a primary reason to run the strategy there.

Don't use Sweden as a standalone strategy. The 131.3% down capture makes it unsuitable regardless of the nominal CAGR.

The strongest portfolio use case: run this screen on India and the US simultaneously. India provides the highest excess returns. The US provides stability, lower drawdowns, and a larger stock universe. The combined portfolio captures the geographic diversification while retaining the behavioral edge.

Part of a Series

Individual market analyses with year-by-year breakdowns:

• 52-Week High Proximity on US Stocks - 12.78% CAGR, 65.0% down capture
• 52-Week High Proximity on Indian Stocks - 26.73% CAGR, the standout
• 52-Week High Proximity on Korean Stocks - 11.43% CAGR, 39.1% down capture
• 52-Week High Proximity on Thai Stocks - 12.55% CAGR
• 52-Week High Proximity on Japanese Stocks - 8.89% CAGR, modest excess

References

• George, T. J. & Hwang, C.-Y. (2004). "The 52-Week High and Momentum Investing." Journal of Finance, 59(5), 2145-2176.

Run the global screen: https://cetaresearch.com/data-explorer?q=JfyjnAI7cd

Data: Ceta Research (FMP financial data warehouse). 18 exchanges tested, quarterly rebalance, equal weight, 2000-2025. Returns in local currency vs SPY (USD).