Subscribe to the 'Sharp Investing' newsletter free!
Issue XI

March 1997

Sharp Investing

Selling Strategies: When to Sell a Winner

Does the following sound familiar? You have a stock that doubles within a short amount of time. You carefully evaluate the stock and decide that it is overvalued, so you sell. Immediately after selling, this stock doubles again, and then again, and then again... We have all been in this situation and 20/20 hindsight provides a harsh measure of our selling strategy imperfections.

It is often said that knowing when to sell a stock accounts for most investment success. While I don't believe this is entirely true, there is no doubt that the decision to sell is the most difficult decision to make in the investment process. The difficulty for investors is that selling a stock at exactly the top is virtually impossible, yet the sale is always judged critically after the fact by using perfect 20/20 hindsight.

In this issue of Sharp Investing, we will discuss successful selling strategies. These combine both art and science to help investors maximize their returns over long periods of time.

Successful selling is NOT defined by "selling at the top." A good example: My father, Robert Sharp, has compounded money at over 30% annually over the last 37 years, yet has only succeeded at selling exactly at the top one time out of a total of over 300 trades.

Selling at the exact top is completely due to luck, nothing more. Short term movements of the stock market and of individual stocks are completely random. Therefore, selling strategies should focus on long run results and should not be compared against the impossible standard of 20/20 hindsight. If investors take this long run attitude in their selling strategies, the emotional issue of selling at the "wrong time" in the short run is minimized.

More Investors that get rich do so by selling TOO SOON rather than too late. This cliché is reinforced by the Robert Sharp example. In spite of his outstanding investment performance over a very long period of time, Mr. Sharp can point out dozens and dozens of stocks that in hindsight he sold much too soon. It is fairly simple to tell when a stock becomes overvalued, but very difficult to determine how much more overvalued it will become before returning back towards its fundamental economic value.

This growth or momentum is heavily relied upon by "growth" investors. They rely on the continuing momentum of the market or a particular stock, having to decide when to hop off the wave before it crashes down on them. Momentum investing, therefore, pays no attention to underlying economic value of the investment, no matter how far overvalued the stock becomes. Since any selling hinges on short term behavior of the market or a stock, 99% of investors will either get out too early (as most value investors do) or too late (as most growth or momentum investors do).

What then, can an investor do to improve the odds of being right in the long run when selling an investment?

1. What goes up must come down
There is an academically verified phenomenon at work in the stock market known as mean reversion of stock market and stock returns. Simply stated: Stocks which have produced high returns over the previous few years are more likely to produce low returns for the next few years. Conversely, stocks which have produced low returns over the previous few years are more likely to produce high returns over the next few years. Of course, there are many exceptions where stocks produce high (or low) returns for very long periods of time (over 10 years). However, on average most stocks exhibit this cyclical "what goes up must come down" phenomenon, giving investors an opportunity to take advantage of this behavior by incorporating it into selling decisions designed to beat the market. The buy and hold forever strategy, on the other hand, simply matches market returns by design.

My own financial research and that of other academics shows that large stocks exhibit maximum mean reversion after 3 years, and small stocks exhibit maximum mean reversion after 1 1/2 years.
What this means is that the best average holding period for a large stock is 3 years (if you bought it as an undervalued stock) and for a small stock is 1 1/2 years. If you are consistently selling your winners before these average holding periods you are likely giving up some extra return available by holding a bit longer. Conversely, if you consistently hold stocks longer than these holding periods you may be giving back some of the extra return by holding too long.

2. Selling based on general market conditions
It has also been academically verified that selling winning stocks in a generally overvalued market provides better investment returns than selling winning stocks in an undervalued market. Of course, there are always differences of opinions on whether a market is overvalued or undervalued, but it is usually a fairly apparent condition at the extremes. The old adage of "buy low sell high" works much better than "buy high, sell higher".

3. Selling based on historical valuations
Another strategy verified by history is selling a stock when it reaches extremes in valuation based on its own history ,(i.e. when it is selling at the high end of it's historical Price/Earnings range, Book/Market range, Sales/Price, etc.). This must be balanced against any fundamental changes in the company itself. If there are no major differences in the company's prospects today versus its past history an investor has a good chance of maximizing returns by selling this holding at the historical top of its valuation cycle. However, if a company has truly "reinvented" itself, historical levels of valuation may not apply. The key determination is the difference between the perception that "things are different this time" and the reality that most companies (especially large ones) do not truly reinvent themselves even though they would like the investing community to believe that they have.

4. Fundamental changes in the company
In some cases, the initial reasons that an investor purchased a stock are no longer valid because of changes in the company. Companies merge, spin off divisions, acquire other companies, and are acquired by other companies. If the new entity doesn't fit your initial criteria for purchase, it may be time to sell your winner. Other fundamental changes could include companies discontinuing certain products or services, starting new products or services, or drastically changing their business strategy. All major changes need to be evaluated by investors to make sure your reasons for being a part owner in this particular business are still valid.

5. Selling based on underdiversification
If an investor starts with 10 stocks, it is not uncommon a few years down the road for a few big winners to dominate the portfolio, sometimes turning into 50% of the total portfolio because of their large gains. While allowing big winners to dominate your portfolio is not a bad strategy, at some point an investor can become overweighted in a particular stock or industry and therefore take on unnecessary risk by becoming under-diversified.

Statistically, anytime a single stock holding becomes more than 1/3 of your total portfolio it makes sense to sell off enough of your big winner to remain adequately diversified. Of course this 1/3 rule can be adjusted either way dependent on the nature of the big winner. There is less risk in holding more of a big, safe, stable stock and conversely much more risk to holding a large amount of a smaller, more volatile stock. Of course, winning stocks that have a big run-up are much more likely to be the former rather than the latter.

6. Tax Considerations
When selling a winner has capital gains consequences, an investor needs to make sure the reasons for selling outweigh the tax liability. In many cases, it is wiser to take the
chance that the market may take back some of your high returns rather than guaranteeing your investment loss to Uncle Sam in the form of untimely taxation of your gains.

7. When stocks exhibit negative non-random short term behavior
When winning stocks exhibit negative short term returns above and beyond their historical short term volatility it can either be viewed as a buying opportunity or a selling signal. Long time readers of this newsletter will recall that several years ago I did a lot of research on trying to determine if the non-random short term negative movements could be used successfully as either a buy or sell signal (applying Statistical Process Control to investment returns). The results of this research are still inconclusive.

Currently there is no historical data that will allow me to separate negative movements based on publicly available information from negative movements based on privately held information. My theory is that negative movements on publicly available information are generally buying opportunities while negative price movement on privately held information are generally selling opportunities. I am working on a real time study that will provide some verification of this theory. Until then the "no news is bad news" theory is just that - a theory.

Sharp Investing is a quarterly publication focused on investment education. For a subscription contact
Sharp Investments, at:

13160 SW Butner Road
Beaverton, OR 97005
Phone 503-520-5000

Fax 520-0530 email:
Daniel R. Sharp
Registered Investment Advisor

8. Tiebreakers
If the above seven strategies fail to provide a clear cut decision in regards to whether a winning stock should be sold or not, the following tiebreakers can be applied.

  • The King Solomon approach - sell half and keep half
  • The principal out approach - sell your initial investment in the winner and continue to ride your winner with the market's money
  • The limit order approach - set a sell order at a premium to the current price and let the market decide whether you sell the security or not

The bottom line - selling a winning stock should be a result of weighing the various reasons for selling versus holding by using a checklist, and recognizing that whatever your decision, you will seldom be right in the short run.

Sharp Investments
13160 SW Butner Road
Beaverton, OR 97005

Selling a Winner Checklist

How close to the 3 year/1 1/2 year optimal holding period is the stock?

Is the stock market overvalued or undervalued?

Is the stock at the high end of it's historical valuation cycle?

Is the historical valuation cycle still valid or is the company different than in the past?

Will not selling the winner expose the portfolio to under-diversification?

Is the tax liability of selling an important factor?

Are the reasons for purchasing the stock in the first place still valid?

Has there been recent declines in price not explained by publicly available information?

Tiebreaker Options

· Sell half, keep half
· Sell principal, keep gains
· Set a limit order to sell above current price

Knowing when to sell a losing stock is just as important to investment returns as knowing when to sell a winner. This topic will be addressed in a future issue of Sharp Investing.

Subscribe to the 'Sharp Investing' newsletter free!