As of 12/05/2024
  Indus: 44,766 -248.33 -0.6%  
  Trans: 16,976 -190.93 -1.1%  
  Utils: 1,047 +2.22 +0.2%  
  Nasdaq: 19,700 -34.86 -0.2%  
  S&P 500: 6,075 -11.38 -0.2%  
YTD
 +18.8%  
 +6.8%  
 +18.8%  
 +31.2%  
 +27.4%  
  Targets    Overview: 12/02/2024  
  Down arrow44,000 or 46,000 by 12/15/2024
  Down arrow17,025 or 18,000 by 12/15/2024
  Down arrow1,025 or 1,100 by 12/15/2024
  Up arrow20,000 or 18,500 by 12/15/2024
  Up arrow6,200 or 5,900 by 12/15/2024
As of 12/05/2024
  Indus: 44,766 -248.33 -0.6%  
  Trans: 16,976 -190.93 -1.1%  
  Utils: 1,047 +2.22 +0.2%  
  Nasdaq: 19,700 -34.86 -0.2%  
  S&P 500: 6,075 -11.38 -0.2%  
YTD
 +18.8%  
 +6.8%  
 +18.8%  
 +31.2%  
 +27.4%  
  Targets    Overview: 12/02/2024  
  Down arrow44,000 or 46,000 by 12/15/2024
  Down arrow17,025 or 18,000 by 12/15/2024
  Down arrow1,025 or 1,100 by 12/15/2024
  Up arrow20,000 or 18,500 by 12/15/2024
  Up arrow6,200 or 5,900 by 12/15/2024

Bulkowski on Stock Relative Strength

Part I

This study is the first part of two that discusses stock relative price strength, whether stock performance when compared to its peers is a good indicator of future results.

Summary
Methodology
Results
25 Stocks
100 Stocks
Length
Trading
See Also

Stock Relative Strength: Summary

In a bull market, select stocks with high price relative strength (strong upward price momentum). They rise quickest. However, in a bear market, avoid the top ranked stocks because they fall fastest. Select those with mid-range relative strength. They decline least. After a bear or down market, switch back into the top ranked stocks.

Stock Relative Strength: Methodology

I measured the percentage change from today's close to the close six months ago and then ranked the results. I did this for each stock (up to 508 of them), beginning from 1/1/1995 until 12/24/2007. That means, the day after the January 1995 New Year's holiday, I looked back six months and computed the price range, sorted, and saved the rankings. Then I looked at the next day and made the same computation. I continued this method until reaching the end date.

After building the database, I used two basic periods of study, the bear market in the S&P 500 index from 3/24/2000 to 10/10/2002, and the bull market thereafter, from 10/11/2002 to 12/21/2005. I stopped at 2005 because I looked two years into the future to see how the stocks behaved and that would take us to the end of the study period.

The easiest explanation of how I did this comes from looking at the results. The below table shows the average performance of 10 stocks from three ranked groups: the best ranked stocks (1-10), middle (ranks 224-234) and worst (448-458) for the bear market. Only 458 stocks existed in the database back then.

I looked at each of the 10 stocks and found the average price change for each day in the measurement period (from 3/24/2000 to 10/10/2002) to a date from one month to two years into the future. For example, if 3M had a relative strength rank of one on 3/24/2000, then I measured the change in price from 3/24/2000 to one month later. Then I did the same for the second ranked stock then the third and continued until the tenth ranked stock. Then I advanced to the next day, found the top 10 ranked stocks, and measured the future price move. I continued this procedure until I reached the end of the bear market. Then I averaged the results and found that they showed a decline of 4% over the month and at the end of that period, the average rank had dropped to 26.

Then I moved to the next period, two months, and measured the average price change of the ten stocks over the coming two months for each day of the bear market. They showed that the decline had worsened to a loss of 5% and the average rank had dropped to 57. I continued this procedure until finding the price change two years in the future for each day in the bear market.

Top of page

Stock Relative Strength: Results

The following table shows the average price performance and average rank of 10 stocks in the bear market. This shows that the top ranked stocks tumbled the farthest followed by the worst ranked stocks. The mid range ones held up the best.

During the two year test period, the best-ranked stocks tumbled 31%, the worst tumbled 11% and the mid range ones actually climbed 3%. Since I used each day in the bear market and looked forward from there, you can probably think of the results as the performance of a stock midway (about 15 months) into the bear market projected one month, two months, and so on, into the future. Coming out of the bear market is why you see a recovery from -2% to 3% for the middle range, and an improvement in the worst ranked stocks (but the best-ranked ones were still dropping).

10 stock range, bear market from 3/24/2000 to 10/10/2002.

1-10Avg224-234Avg448-458Avg 
BestRankMiddleRankWorstRank
-4%260%225-5%4341 month
-5%57-1%226-9%4122 months
-7%100-1%224-12%3823 months
-18%228-2%227-15%2546 months
-28%250-3%231-19%2469 months
-30%233-2%231-22%2371 year
-31%2483%238-11%2092 years

Top of page

Notice how the average rank drops from 26 to 248 for the top ranked stocks. The middle group shows a stable rank and the group with the worst relative strength shows the average relative strength rank improving form 434 to 209. In other words, the top ranked stocks took a big hit during the bear market. Those in the middle suffered, but not nearly as much, and the worst ranked stocks showed improving relative strength.

 

10 stock range, post bear market, from 10/11/2002 to 12/21/2005

1-10Avg231-241Avg462-472Avg
BestRankMiddleRankWorstRank
4%191%2382%4321 month
7%413%2444%3882 months
10%724%2476%3483 months
19%1918%25514%2156 months
25%20412%25221%2169 months
32%22516%24727%2191 year
52%21832%25143%2582 years

Top of page

This table shows the average performance of 10 stocks after the bear market. The number of stocks increased to 472 because those are the ones that existed in 2002. The results show that the best-ranked stocks had the highest improvement over time. The middle group was the worst performing and the stocks with the lowest rank came in second.

For example, the top 10 ranked stocks for relative strength soared an average of 52% over the two-year test period. The stocks with the worst relative strength climbed 43%, and the middle group climbed just 32%.

Stock Relative Strength: 25 Stocks

The prior two tables used 10 stocks, but the next two tables use 25 stocks covering the same period, a bear market and then a bull market. The results are similar. During a bear market, avoid those stocks having the highest relative strength. The middle group does best.

25 stock range, bear market from 3/24/2000 to 10/10/2002

1-25Avg211-236Avg433-458Avg
BestRankMiddleRankWorstRank
-2%410%223-3%4211 month
-3%77-1%223-6%3942 months
-6%117-1%222-8%3663 months
-16%237-2%226-13%2516 months
-23%256-3%231-16%2419 months
-25%239-3%233-18%2361 year
-23%2473%239-12%2332 years

Top of page

The next table shows the results of the bull market study. The stocks with the highest relative strength perform best followed by those with the worst relative strength. Stay out of the middle group because they perform worst.

 

25 stock range, bull market from 10/11/2002 to 12/24/05

1-25Avg217-242Avg447-472Avg
BestRankMiddleRankWorstRank
4%341%2362%4211 month
7%613%2414%3792 months
10%914%2456%3413 months
19%1968%25513%2216 months
25%21512%25320%2199 months
31%23816%24925%2231 year
54%22132%25140%2632 years

Top of page

Stock Relative Strength: 100 Stocks

The next two tables are the same as the prior ones except they use 100 stocks. I just wanted to be sure that the results applied to a large sample of stocks, not just a handful. The worst performers are the ones showing the best relative strength at the start of a bear market. The ones in the middle of the 458 stock list retain their value best. In second place are those with the worst relative strength.

Since I looked at the performance from the start of the bear market, it may be that the performance profile is different if started from the middle of the bear market, so keep that in mind. In the below table, notice that the rank drops dramatically in the best column, are stable in the middle column, and improves for those with the worst relative strength.

 

100 stock range, bear market, from 3/24/2000 to 10/10/2002

1-100Avg154-254Avg358-458Avg
BestRankMiddleRankWorstRank
-1%890%210-2%3741 month
-2%1230%214-4%3462 months
-3%153-1%217-5%3193 months
-9%240-3%227-8%2386 months
-12%244-3%231-10%2349 months
-13%239-3%232-11%2361 year
-10%2363%238-5%2392 years

Top of page

In a bull market, find stocks with the best relative strength and buy them, as the below table shows. For the other two categories, middle and worst, performance is similar.

 

100 stock range, bull market from 10/11/2002 to 12/24/2005

1-100Avg186-286Avg372-472Avg
BestRankMiddleRankWorstRank
2%841%2392%3821 month
5%1133%2433%3502 months
7%1394%2475%3223 months
14%2209%25410%2406 months
17%23512%25216%2339 months
21%25016%24921%2321 year
44%22832%25134%2622 years

Top of page

This last table (below) shows the performance for the entire span from 1/1/1995 to 1/21/2005. After two months, stocks with the best relative strength do best until the two years category. At that time, all of the columns reach 20% average gains.

 

100 stock range, all years from 1/1/1995 to 1/21/2005

1-100Avg141-241Avg282-382Avg
BestRankMiddleRankWorstRank
1%821%2011%3041 month
2%1102%2082%2852 months
4%1352%2132%2703 months
7%2095%2284%2316 months
9%2187%2296%2319 months
11%2309%2309%2271 year
20%22820%23220%2362 years

Top of page

Stock Relative Strength: Length

How long does a stock stay top ranked? I looked at all stocks in the database from 1/1/2005 to 12/24/2007 and counted the length of time that a stock held each rank. A contiguous run ended if the same rank did not appear for a week or longer.

The results appeared saucer like - higher on the ends than in the middle. The top rank of one occurred 252 times and remained top ranked for an average of 20 days. Rank two: 10 days. Rank three: 7 days, and at rank nine, the average drops to 3 days. At rank 16, we are into the 2s. At rank 65, we are into the 1s, meaning stocks ranked 65 lasted one day at that rank. The one day average continued until rank 363 where the average ticks up to 2 days. Solid twos begin at rank 387 then threes take over at 469, fours at 494, eventually ending at ten days for rank 504, 16 for rank 505 and an average of 17 days for the worst relative strength rank of 506. The worst rank had just 14 samples. Incidentally, stocks with an average of 1 day per rank had about 3,000 samples.

The numbers suggests that the middle ranks are fluid, changing ranks from day to day. At the ends of the list, it appears that the very best continue to perform well and the very worst continue to have problems, at least for a few weeks.

Since I ranked the stocks over a weekend, stocks splits, dividends, spin-offs and such had no effect on the database.

Top of page

Stock Relative Strength: Trading

If you are in a bull market, select stocks with the highest relative strength. Even in a list of 472 stocks, using the first 100 gives better performance than lower ranked ones.

If you are going into a bear or down market, then avoid those stocks with high relative strength. They crash and dig a deep hole. You'll want to avoid stocks with the worst relative strength, too, so select from the middle of the list.

When the market resumes the uptrend, stocks having the best relative strength recover first followed by those with the worst relative strength. In last place are the middle group. Either trade stocks with the best or worst relative strength and stay out of the middle.

Go to part two

-- Thomas Bulkowski

Top of page

See Also

 

Support this site! Clicking any of the books (below) takes you to Amazon.com If you buy ANYTHING while there, they pay for the referral.
Legal notice for paid links: "As an Amazon Associate I earn from qualifying purchases."

My Stock Market Books
My Novels

Copyright © 2005-2024 by Thomas N. Bulkowski. All rights reserved.
Disclaimer: You alone are responsible for your investment decisions. See Privacy/Disclaimer for more information.
Some pattern names are registered trademarks of their respective owners.
Home Advertise Contact Privacy/Disclaimer

Love may be blind, but marriage is a real eye opener.Smiley