The Outlier Effect

1/25/2016


Like most well-worn Wall Street maxims, "let your winners ride" contains more than a grain of truth.

Because so much of the typical portfolio's return derives from its biggest winners, investors in a rush to lock up profits can cost themselves dearly. Portfolios selected using our Quadrix rating system are heavily influenced by this outlier effect, especially when they include small and midcap stocks.

To illustrate, I looked at the performance of six hypothetical portfolios, all based on stocks in the broad Dow Jones U.S. Index since 1992. Portfolio No. 1 selected all stocks in the index with Overall Quadrix scores of at least 90 and held for 12 months. In the 276 rolling 12-month periods ended December, this buy-and-hold portfolio delivered an average return of 16.7%, as shown in the table below.

The second portfolio was constructed the same way, but on a monthly basis it sold any stock with a gain or loss of 10% or more from initial purchase price. While this approach lowered portfolio volatility considerably, it affected returns even more. Average 12-month return was just 5.7%.

Portfolio No. 3 waited for a move of 20% before selling. But the drag on returns was still dramatic, with an average 12-month gain of 11.4%.

If 20% exceeded the average return of the stocks, why did selling 20% movers crimp returns so much? The answer lies in the results of portfolio No. 4, which held all stocks for 12 months but excluded the 5% with the best returns. On average, this meant we excluded six out of 166 stocks from the calculations. Portfolio No. 4 returned 12.0%, on average.

Interestingly, portfolio No. 4 had a lower standard deviation than portfolio No. 1, as the volatility it eliminated was the good kind. On average over the 276 rolling 12-month periods, the biggest winner in No. 4 was up 95.7%. For No. 1, the average gain of the biggest winners was 222.5%. While huge gains in the tech boom of the late 1990s skewed the numbers, in one-half of periods the biggest winner returned at least 165.9%.

To be sure, excluding the worst losers also has a dramatic impact, lifting the average return to 21.3%, as shown in our fifth portfolio. But selling any stock with a 20% move does not do a very good job of limiting losers. And even if you had perfect foresight and could exclude both the top 5% and the bottom 5% of performers, as shown in portfolio No. 6, you wouldn't match the return of No. 1.

Lessons for investors

Obviously, I'm presenting six highly hypothetical portfolios, and back-tested returns prior to 2000 benefit from the hindsight I've incorporated into Quadrix over the years. Also, I'm assuming stocks sold earn 0% returns for the remainder of the 12 months. Still, the math overwhelmingly suggests you shouldn't sell just because you are up or down 20% in a stock.

Once you own a stock, the price you paid for it is irrelevant except for the tax implications. In general, taxes favor selling losers and keeping winners. That does not mean you should hold on to second-rate stocks to avoid taxes. Nor does it mean you should chase richly valued stocks simply because they have performed well. It means you should be open to the idea that some of your best picks may be stocks in which you already have a big gain.

The high price of locking up profits
-------- Six Portfolios, All Based On Stocks With Overall Quadrix Scores Of At Least 90 --------
1. Hold All
Stocks For
12 Months
2. Sell Any
Stock If Up
Or Down 10%
3. Sell Any
Stock If Up
Or Down 20%
4. Hold All
But Exclude
Top 5%
5. Hold All
But Exclude
Bottom 5%
6. Hold All
But Exclude
Top And
Bottom 5%
Average 12-month return (%)
16.7
5.7
11.4
12.0
21.3
16.5
Standard deviation (%)
19.5
8.0
13.2
18.5
20.1
19.0
Best 12-month return (%)
99.8
24.8
44.0
88.7
108.4
96.7
Worst 12-month return (%)
(38.8)
(21.2)
(24.6)
(40.3)
(36.0)
(37.5)
% of 12-mo. periods up (%)
84
78
84
79
90
85
Average number of up stocks
105
102
105
99
105
99
Average number of down stocks
61
64
61
61
50
50
Average gain of biggest winner (%)
222.5
63.6
110.7
95.7
222.5
95.7
Average loss of biggest loser (%)
(70.9)
(41.8)
(55.1)
(70.9)
(34.4)
(34.4)
Notes: Based on 12-month total returns for stocks in the Dow Jones U.S. Index, since Jan. 1992. Standard deviation measures the dispersion of 12-month returns, with a higher number implying greater volatility.

 


Current Hotline

Stock Spotlight

Individual Stock Reports

ISRs make stock research easy!

Perhaps the most valuable two page reports available anywhere.

All the data you would normally have to plow through years of 10-K filings, earnings reports, and reams of market data to assemble — yours all in one concise report.

ISRs contain our proprietary Quadrix scores — find out how we rate all the stocks in the S&P 500.

Visit us at individualstockreports.com