Since this web site is read by many very successful investors, I was wondering if any of you had any insights on selling stocks. It seems there are a million different books and strategies on buying stocks, but not many people talk about when to sell. Right now I basically go off of what I am satisfied in gaining and start selling at that point. If it drops I usually keep buying as long as my analysis of the company still holds. This has worked out okay but I've missed out on a lot of gains and feel there is more to learn here. For instance, I bought ZINC at $2.50. It went up and bounced around from $3 to $4 for a while and I determined that at $3.50 I was satisfied with a 40% gain. Of course after I do this it shoots up to over $7 over the next couple months. On the other hand, in an attempt to exercise more patience, I held on to TRID when it was up over 40% and I am now in the negatives with it. I currently am up 110% on TUES and am confused about what to do with it. My instinct has been telling me to sell and lock in the gains but it just keeps going up. My dad once told me to "Sell and be sorry, but sell." This seems like pretty good advice but I can tell you when a stock keeps rising I sure am sorry I sold. Could anyone help me out? Not sure if it matters, but my investing style is that of value. Thanks in advance!… Brave Rifles!

Philip J. McDonnell replies:

Phil MMr. Bates posed the interesting question of when to sell. Much of his discussion focuses on minimizing his subsequent regret. Many of us buy stocks because we fear the train will leave the station without us on board. We fear taking a loss because we cannot accept the cognitive dissonance of admitting a mistake. We also fear that the stock will recover from the loss adding regret to the cognitive dissonance. We fear selling a profitable trade because it may go higher and cause us regret. For most counters the question of when to sell is often answered before they buy. Suppose one did a study such as Vic and Laurel did in PracSpec looking at REIT returns in one quarter and what the market did in the next quarter. Then the criteria of when to sell is one quarter later just as used in the study. In fact very often the criteria used in a quant study is based on time and not on price. One can suggest a few general guidelines. One should sell if:

  1. The criteria used in your analysis have been met. This could be time or it could be something like the first profitable open such as Larry Williams has suggested.
  2. You no longer have an advantage.
  3. You have a profit so large that the position size is too large relative to your portfolio. Note that this criterion only requires a partial sale. My book talks about position sizing.
  4. You have other better opportunities and need to raise cash.

For a fundamentally oriented value investor the time element may not be so clear. Corporate reporting is on a quarterly basis so presumably the time frame for a value investor is quarterly or longer. So one should look to value as the criterion for exit as well as entry. If a stock has gone up quite a bit then much of its advantage in the value sense has disappeared. The position size may now be larger than the optimum and other under valued stocks may present better opportunities.

Correct position sizing is the only reliable money management tool. Stop losses are not guaranteed to work because of gap openings. They also do not work the way most people expect. Buying puts is usually too expensive. Thus we are left with position sizing. For a value investor with longer holding periods one might arbitrarily decide to cut positions in half to allow room to double before position size adjustment is required.

Dr. McDonnell is the author of Optimal Portfolio Modeling, Wiley, 2008


WordPress database error: [Table './dailyspeculations_com_@002d_dailywordpress/wp_comments' is marked as crashed and last (automatic?) repair failed]
SELECT * FROM wp_comments WHERE comment_post_ID = '3709' AND comment_approved = '1' ORDER BY comment_date




Speak your mind


Resources & Links