In this paper we tested the hypothesis that the performance of the best technical trading rule is no better than that of the buy-and-hold strategy, uSing White's Reality Check and Hansen's Superior Predictive Ability test. We find that the null hypothesis was not rejected in the Korea Stock Exchange (KSE) market when the investment performance is measured with the simple return , but rejected based on the Sharpe ratio criterion. Therefore, we cannot be confident that the technical trading produces economically valuable signals in the KSE But in the KOSDAQ market the technical trading showed better investment performances on both evaluation criterions , which implies that the investment performance of the technical trading rule comes from its predictive ability, not from the simple luck. Empirical studies on the performance of technical trading applied to the stock price indices of small , medium and large-capital size show that the investment performance of the technical trading rule was not reliable , irrespective of the capital size, when the performance is evaluated with the simple returns. But the investment performance on the small-capital size was statistically trustworthy