Technical Trading Strategies: Profitable or Just Curve-Fitting?

Technical analysis (TA) is a trading approach that evaluates past price movements and volume data to forecast future market trends. It relies on price patterns, indicators, and statistical measures rather than fundamental factors. Investors use tools like moving averages, RSI, and Bollinger Bands to identify trends, support and resistance levels, and potential entry or exit points.

The growing proliferation of academic articles on technical analysis suggests its acceptance and effectiveness. However, some researchers continue to question its validity. Reference [1] revisits the question of whether TA works, testing TA rules on three stocks—AAPL, MSFT, and NVDA—from January 2000 to December 2022. The study evaluates the profitability of various technical trading strategies both in-sample and out-of-sample using methods such as reality checks and stepwise tests.

The authors pointed out,

In this paper, we analyze a comprehensive dataset of AAPL, MSFT, and NVDA stocks from January 2000 to December 2022 to evaluate the profitability of various technical trading strategies both in-sample and out-of-sample, using February 2016 as the primary cutoff and May 2018 as an alternative. We construct strategies based on multiple indicators and timeframes, conducting thorough statistical analyses to ensure robustness against data-snooping bias. Our results consistently demonstrate that apparent profitability often stems from parameter selection rather than true market inefficiencies, supporting the efficient market hypothesis. This highlights the difficulty in predicting profitable strategies ahead of time, emphasizing the unpredictable nature of achieving sustained trading success.

In short, the article concludes that TA does not work, as it fails to identify any technical trading strategies that yield consistent profits across both periods. The results consistently show that apparent profitability often arises from parameter selection rather than genuine market inefficiencies, supporting the efficient market hypothesis.

We welcome this type of research that challenges prevailing beliefs. However, in our opinion:

• The sample size is small, covering only three stocks,

• While the examined period is long, market dynamics may have changed over time. Using an extended dataset is beneficial, but a trading system should account for shifts in market conditions,

• Only two types of tests were conducted; more comprehensive testing is needed.

That said, we look forward to seeing further research in this area. Let us know what you think in the comments below.

References

[1] Wang, Y., Chen, Y., Tian, H., & Wayne, Z. (2025). Evaluating Technical Trading Strategies in US Stocks: Insights From Data-Snooping Test. Journal of Accounting and Finance, 25(1).