“How do I know when to sell a stock?” That’s a common question that most investors struggle with and one that I am asked quite frequently. One technique that I use to change signals is “expected value” - a concept borrowed from basic probability theory that is defined as the sum total of all possible outcomes times their related payoffs. Simply put, if I think a stock has a high probability of going up but the expected gain is relatively small while the risk to the downside is high, I will make an exception to my normal procedures and change the signal to Down before I have any firm evidence in the form of weakening fundamentals or technicals. The most frequent situation that benefits from this kind of analysis is when a stock has gone on a nice run and you start having that nagging fear that you should get off the joyride before it gets ugly. Some people call it taking money off the table or profit taking, but I just call it a normal part of portfolio management. So even though it’s impossible to get all the probabilities and payoffs perfectly thrown into a formula, I recommend that you do some quick expected value calculations in your head and keep a close eye on your other fundamental and technical factors.