To: Elroy who wrote (54708 ) 1/4/2015 1:56:50 AM From: Jurgis Bekepuris Respond to of 78627 To answer your question: people make fun about analyst "hold" recommendations, but there is such thing in real life. :) There are actually three somewhat overlapping areas of stock price: undervalued, fairly valued, and overvalued. "Fairly valued" is not a point on a line. It is a (possibly large) interval. In your example, $0-$30 is undervalued, $31-$45 is fairly valued (I just made up $45 for example's sake) and $45-$10000 is overvalued. :) You don't want to buy stock that is fairly valued and you don't want to sell it there, but you want to hold it if you have it. So here's where the "hold" recommendation fits. :) If you look at Buffett, even though he says he's "sell never" investor, he does sell some things. But his "fairly valued" interval is very big, so he usually does not sell at all.I've always found the decision of where to sell a lot harder than where to buy. My sell strategy, to the extent that I have one, is don't sell losers. It leaves me with a catch-22. I tend to not sell the losers. And the winners, well they usually have improving stories and are doing well, so I don't want to sell them either! Somewhat agreed. There are some pointers for selling though: - If significantly better opportunity comes along. This requires comparative evaluation of two opportunities that may be hard. (Very hard for me usually) - If business deteriorates materially and it's not clear if it will ever recover. (Unless the stock dropped 90%+ already - sell, if it dropped 95% - think hard and maybe still sell). - If stock runs up way above fairly valued. (But it could run up even more... see MA, V, GOOGL, AAPL, etc... but it could crash back too, see LVLT in 2000 :)). Some people make the following suggestions: - Sell your dogs. "Don't watter your weeds". IMHO stupid if the company crashed temporarily - like you said, you could buy more and make tons of money. You need to sell the stock if company is not doing well and might not do well in foreseeable future though. - Take some money off the table, i.e. sell some of the winner. Might be stupid too if you have BRK, MSFT, AAPL or whatever long-term winner company. I did some of this though, since most of the time the company is not BRK/MSFT/AAPL. - Look at the stock as if you were buying it. If you would not buy it here, sell it - again about selling winners. This is stupid, since it ignores the fact that there is a large "fairly valued" interval. Perhaps better way to think about it is: what was your goal/expectation/plan? If the expectation was to hold stock forever, then perhaps you should hold it forever whether it falls or rises. If the expectation was that a stock was net-net cigar butt and it should recover to 1.2x book, then sell at 1.2x book even if it seems that it might run to 2x book. If the expectation was to sell the stock when oil recovers to $97, then sell when oil goes to $97 even if you think that it might go to $192. Or at least reevaluate hard. :) It is possible to have similar plans for drops: if oil drops to $30, I will add 2x, but not more. If oil drops to $10, I will not add more. I think this might be worth doing beforehand, though few people do it and even fewer follow their plans... :)