Do You Buy Breakouts? Think Again (Part 2)
Monday December 15, 10:00 am ET
By Larry Connors
Further Stunned!
I thought I had gotten over the findings from last week's results of buying breakouts but then we ran more tests and I'm now even more stunned!
A Brief Review Of Last Week's Findings
We now know that had we bought new 10 day highs in the S&P's and exited when prices crossed below their 10-day ma, we would have lost money in the bull market of the past decade. We also know that had we acted counter-intuitively and sold when everyone else was buying at these times, we would have made good money. And most importantly, as the world was throwing in the towel and grumbling about how bad things were and how "the internals" were deteriorating, if we had been buying during these times (at 10 day new lows) and had sold when prices rose above 10-day ma, we would have far outperformed buy and hold plus we would have been in the market less than 40% of the time. A nice combination. Wow, buying low and selling high...what genius!
Now Let's Move To Other Markets
Let's now go further. Let's look at some other indices. First the Nasdaq. The market that had one of the greatest runs in US history in the late 90's and then one of the greatest collapses. I'm going to first show you the annual results for the past 11 years. The column on the top shows us the number of Nasdaq points that were made by buying a new 10-day low (buying the pullback) and exiting when prices crossed above the 10-day moving average.
Buy The 10-Day Low -- Exit When Prices Close Above The 10-Day Moving Average
Total Nasdaq points gained: 2940.85. Percentage of Trades Profitable: 75.1% (does not include slippage and commission). And, you accomplished all this by being in the market less than 32% of the days during this period. Yes, that's right, 68% of the time your money was not at risk. It was in cash.
The column on the bottom shows what happened if you only traded the breakouts -- you plowed in when everyone else was buying and jumping up and down about how great things were. You bought the 10-day highs and exited when the market crossed under its 10-day moving average.
Doing The Opposite -- Buy The Breakouts -- 10 Day New Highs, Exit is When Prices Close Under The 10-Day Moving Average:
Total Nasdaq points LOST: -840.05. Percentage of Trades Profitable: 33.5% (does not include slippage and commission).
Want More?
Let's look at the Semi's (CBOE:^SPX - News). Same rules. Buy the 10-day lows and exit above the 10-day ma. Total gains were 1132 points and you were correct 74% of the trades. Buy the breakouts in the SOX? Well, not so pretty -- you made about 1000 points less. Banks? Yes, you guessed it. Pretty much the same results. You made far greater returns in the Banking Index buying 10-day lows vs. buying the 10 day highs. I could go on and on and show you this over and over again and I can do this with different time frames. The results all follow the same path.
Be The Butcher!
1. What first and foremost stands out to me is the fact that these rules strongly lend themselves to the argument that buying pullbacks over the past decade has been far superior than buying breakouts. As I stated last week, buying when the world is selling, and selling when they are all buying, is better than being part of a sheep flock. Just as you want to bet on the butcher when he is taking the lamb to the slaughterhouse, you want to be the butcher here and be selling when the lambs are buying and buying when they are selling.
2. The Nasdaq results for buying the 10-day lows are a bit startling. It should have made money in a rising market. But, it was also successful in one of the most severe bear markets seen (2000-2002).
3. Price behavior is price behavior, and the behavior of the stock market does fall into a rhythm. This rhythm is different than the rhythm you see in currency markets and commodity markets (which tend to trend better). This rhythm is a short-term high-to-low-to-high rhythm that plays itself out over and over again. Overbought to oversold to overbought. Not overbought to more overbought to even more overbought. Yes, sometimes it does the latter for short periods of time, but it does not do this for long periods of time. The above statistics, along with last week's statistics, further prove this out.
4. Am I saying that trend following doesn't work? Absolutely not! Add a trend filter component to this (i.e., a longer-term moving average) and trade in the direction of the trend and you will very likely improve performance.
5. This week's and last week's column are NOT intended to be traded as a system. The information is presented to allow you to look at the effectiveness of buying pullbacks vs. breakouts in the stock market indices over the past decade. Protective stops and proper money management should be implemented in any trades you do.
6. There is no guarantee any of this will continue in the future. Death, taxes and the Patriots winning the Super Bowl this year are the only sure things in life.
7. Maybe the most important: Try not to base your decisions upon the emotions and guesses of stock market analysts on TV, radio and in the press. History has proven that no one has EVER gotten rich buying every time "these smart guys" are jumping up and down because the stock market has just risen for the past two weeks. And certainly no one has created wealth by selling when these same cast of characters are depressed because the market has had a few bad days. Short-term market tops coincide when there have been a number of days of good news and when these guys are buying like crazy and short-term market bottoms occur when the news has been bad and these same people are selling. I wish I was smart enough to have thought all this up myself but it's been around forever. Buy low, sell high -- it makes no difference what the time frame is. It's not only simple, but the statistics we've just seen the past two weeks back this up even further.
Have a great week trading (and feel free to email me at lconnors@tradingmarkets.com if you have any questions)!
Larry Connors
biz.yahoo.com |
