This paper evaluates the performance of technical trading rules using White's Reality Check bootstrap methodology to quantify data-snooping bias and adjust for its effects. The study expands on the work of Brock, Lakonishok, and LeBaron (1992), who examined 26 technical trading rules on the Dow Jones Industrial Average over 90 years of daily data. The authors apply the same data to a much larger universe of nearly 8,000 parameterizations of trading rules, allowing for a comprehensive test of performance across all rules. They find that certain trading rules outperformed a benchmark of holding cash during the period 1897–1986, even after adjusting for data-snooping. However, when tested out-of-sample from 1987–1996, the probability that the best-performing trading rule did not outperform the benchmark was nearly 12%, suggesting that technical trading rules may not have provided significant economic value during this period. The authors also test the impact of transaction costs and short-sale constraints on the performance of technical trading rules, finding no evidence of superior performance in the S&P 500 futures market. The study highlights the importance of addressing data-snooping in financial research and demonstrates the utility of the bootstrap methodology in evaluating the performance of technical trading rules. The results suggest that while some technical trading rules may have performed well in the past, their effectiveness may be limited by data-snooping and other factors. The paper also discusses the broader implications of data-snooping in financial and economic research, emphasizing the need for careful evaluation of empirical results.This paper evaluates the performance of technical trading rules using White's Reality Check bootstrap methodology to quantify data-snooping bias and adjust for its effects. The study expands on the work of Brock, Lakonishok, and LeBaron (1992), who examined 26 technical trading rules on the Dow Jones Industrial Average over 90 years of daily data. The authors apply the same data to a much larger universe of nearly 8,000 parameterizations of trading rules, allowing for a comprehensive test of performance across all rules. They find that certain trading rules outperformed a benchmark of holding cash during the period 1897–1986, even after adjusting for data-snooping. However, when tested out-of-sample from 1987–1996, the probability that the best-performing trading rule did not outperform the benchmark was nearly 12%, suggesting that technical trading rules may not have provided significant economic value during this period. The authors also test the impact of transaction costs and short-sale constraints on the performance of technical trading rules, finding no evidence of superior performance in the S&P 500 futures market. The study highlights the importance of addressing data-snooping in financial research and demonstrates the utility of the bootstrap methodology in evaluating the performance of technical trading rules. The results suggest that while some technical trading rules may have performed well in the past, their effectiveness may be limited by data-snooping and other factors. The paper also discusses the broader implications of data-snooping in financial and economic research, emphasizing the need for careful evaluation of empirical results.