The Brown Lady's Trading Secrets

I don't know who the Brown Lady of Raynham Hall was (or is) but her image reminds me of many aspects of technical analysis. If you look long and hard enough at chart patterns, Fibonnaci numbers, moving averages and a host of other technical indicators, patterns often emerge even where there are actually none.

Swing trading has it's place...especially in an uptrending market. But the truth is that the real $$ is made in the longer term market trend. If you happen to be on the right side of those moves, holding during those short term reversals, chances are that your one gain will exceed the sum of those short term swing trades. That is why I have been developing a system that leans toward the big picture.

Transition to longer term hold can take some time. It helps if you look at a weekly versus a daily 12 month chart. Viewing backtested results for the CMI 2.0 over the last year or so is also a big motivator....

BUY DATE SELL DATE P/L % WEEKS IN TRADE
8/18/09_ 01/16/09 ___ 45__________ 20 weeks
02/10/09 _03/04/09 __26 ___________3 weeks
03/04/09_ 06/08/09 __81 ___________12 weeks
6/19/09 _07/02/09 ___-2.5___________ 2 weeks
07/10/09_ 09/23/09 ___48 ___________10 weeks
09/30/09_ 10/15/09 ____4 ____________3 weeks

Notice the time (in weeks) for each trade. Spooky, I know, but the results speak for themselves.

So before you start searching for that next swing trade set-up, consider this different approach. It may save you time and effort, increase your trading capital, and keep you from seeing shadows and strange figures that are gone as fast as they arrived.

Not So Fast

This is the first time in CMI history that I did not jump all over a trade signal from the CMI. This is mainly due to the fact that I am more interested in the CMI 2.0. I won't buy the QID this time and will let the CMI show me up again although I think the CMI will get whipsawed today and have to reverse this signal.

If I am right...score one for CMI 2.0. If the bear in the left sidebar disappears, you'll know why.

I am putting together one of those 'propaganda type' charts that show where a $10,000 investment in the S&P, CMI, and CMI 2.0 would be after 10 months. This will clearly show that hindsight really is 20/20. We can only wish for foresight that will also be 20/20 in the coming months. (I'll even take 20/40 with corrective lenses.)

Stay tuned for more scary posts. I haven't used up all my Halloween jpg's yet.

QID Trade Signal

The CMI gave the green light to go ahead and buy the QID at tomorrow's open. The QID trades this year have not been too kind but the tide is bound to turn eventually. Interestingly enough the CMI 2.0 (currently under construction) is not committing to any trade just yet. The 2.0 only trades half as much as the CMI so this is no real surprise.

It would have been nice to ride the QID down over the last four or seven days but stronger confirmation is required for the 2.0. This prevents getting stuck in a trade due to a false breakout to either side. A QLD trade is more probable next week if following the CMI 2.0. This should be a good head to head test of both methods.

I will continue to post only signals from the CMI for now until 2.0 is ready to go.

In 2.0 We Trust?

Testing for the new CMI continues. At the moment current testing is measuring risk and reward. It would be nice to be able to find a trading method that yields higher gains but often the cost is a risk that a trader might not want hanging there waiting to rear it's ugly head. Trusting a new trading system takes time. The CMI has proven that it can provide fairly safe returns but some of the drawdowns this year plainly showed that it was out of sync with the market and that another approach should be sought out. Back to the laboratory in other words. Out with the old and in with the 2.0.

The current CMI is taking a rest as you can see by the bull and bear in the left sidebar. The CMI 2.0, in it's present state, is also sidelined.

Looking Forward Using the Past

If you could pick all the significant swing points from this year and had bought the QLD or the QID depending on the direction your total gain for the year would be somewhere around 233%. Not a bad "take" for 16 trades and a 93% win rate.

Several more hours have gone into creating a method to pick these swing points...before they happen. All the data crunching has pushed me away from the short term trends and closer to longer term holds. Long term at the moment looks like two months on average. Back testing the current system (aptly named CMI 2.0) for the first 10 months of this year yields a gain of about 120% give or take after only five trades.



This was not exactly what I was looking for since this type of trading requires alot of faith during intra-trade swings. On the other hand, the CMI was originally created as a replacement for those mutual funds that could not get out of their own way during tough times. Safely generating positive returns for your large trading $'s was the goal all along.

Backtesting is painstakingly slow but the rainy season in the jungle is upon us so there is plenty of times to review those charts with the squiggely lines.

Building the Perfect Beast

I will probably get out of the QLD trade that I bought into last Monday as the CMI 2.0 is giving signals to get out. I have gained just under 10% using this new, untested method.

Testing a new market direction indicator can be tricky since you must follow three steps to get it right:

1. Plan first, then gather data;
2. Decipher data so that it provides useful output (buy/sell signals);
3. Backtest and 'forward' test. Just because it looks impressive historically doesn't mean that it will provide useful output in the future.

Currently, I am somewhere between steps 2 and 3. I will provide more indepth information over the next few days. The estimated time of completion for CMI 2.0 is late November.

In the meantime, the CMI is hibernating which means no trading for now. I will post the updated trading results in the left sidebar after I close the current trade.

Catching Up

In an effort to catch up to the CMI, I had bought the QLD on Monday based on the buy signal received from the new, untested CMI 2.0. I think that this system will 'out shine' the CMI since the signals appear to be more aligned with the market swings while smooth enough to prevent too many false signals.

This method will either put you in the QLD or the QID with less time on the sidelines. I am still working out some of the details that have to do with selling (buying the QID) in a uptrending market and vice versa as these signals are not quite as profitable.

Back testing during a flat year such as 2005 should still be done although I will have to play around with some exponential smoothing of the data in order to generate the correct graph.

I will post the results of this first CMI 2.0 trade when I get the sell signal. The old CMI is holding in a 'no trade' pattern and should be there for a while.

CMI 2.0?

The 'no-trade' zone was the place to be this week. Another down day and the CMI may trigger the signal to buy the QID. We will see what next week brings.

While waiting out a tropical downpour this weekend, I have been working on a 'high-octane' CMI that could potentially blow away the current CMI. The necessary backtesting will need to be done, but if things go as planned, CMI '2.0' could be ready by 2010.

Anyone Can Look Like A Genius

The CMI is now officially in the 'no trade' zone after running the year to date totals up to over 30% for the year. Compare this to the Nasdaq and it doesn't look that impressive but it does make trading easy if you are willing to trust the signals.

The CMI could overtake the market if it times the QID trade correctly on the way back down. With one quarter to go, another 20% is possible but we will have to wait and see. There is no forcing the CMI to make a decision.

Anything seems to work in an uptrending market, unless you are short. Let's see how many Einsteins are left by the end of the year.

Time to Sell the QLD

The CMI is giving the signal to end the QLD trade so the sell signal will take place at tomorrow's open. The CMI has turned around and now is up over 29% for the year. If you can put up with several losing or flat trades, following the system seems to work out quite well.

As I had said in the previous post, the CMI will give you two or three large percentage gains in a year but you have to be willing to trust it. Two seven percent losses during the middle of the year did not do much for the confidence level. Not trusting the signals as you can see from my trading of the CMI signals has not been very profitable.

So for now, I'll sit on my favorite branch high above the swamp and wait for the signal to buy the QID...unless the market continues to favor the QLD side.