Trend following (TF) is a rule-based trading mechanism that taps on the movements of long-term market trend instead of relying on any forecast or external information to decide when to buy and when to sell a stock. Its simple operation is in contrast to complicated prediction methods which typically would try to predict a future trend by analyzing the historical data and may be other factors. TF makes no prediction and it is well known for its simplicity. Although TF is a popular strategy in finance and was implemented in some commercial trading system decades ago, there have not been many studies of TF in computer science. Hence the objective of this paper is to develop a computer simulator in which TF is implemented as a variety of algorithms. Through the algorithms, readers will see how different parameters are chosen, and how these TF strategies perform in a simulation with real-life data. The simulation results show that TF algorithm can gain an average profit of 75.63% of return-of-investment monthly. However, we observed that TF degrades in performance in proportion to the amount of fluctuation of the market trend. This finding is important to the design of technical trading systems. It implies that the fluctuation of market trend should be monitored; when it exceeds a certain threshold the TF trading should be paused to prevent loss. © 2011 ACADEMY PUBLISHER.