We discuss the independence between past and future events in random processes. We warn against two mental errors people tend to make when dealing with random processes -- to expect future events to continue past trends and to expect future events to differ from past trends.