As we are currently sitting in a significant drawdown, it is important to revisit the reasons that we have chosen to not trigger a sell signal weeks ago and to look at what other market timers have been doing.

When we first designed our systems we knew that it was inevitable to have large drawdowns. If we tried to avoid this inevitability, we would have ended up with many more trades and even larger drawdowns. We originally had to decide between two different ways of dealing with corrections. The first option was to employ a trend following system that would move a buy signal to sell when certain technical conditions were met. Using this, we would have had a sell signal on May 15^th and we would still be in sell mode. The NASDAQ Composite has decline around 10% from there. The second option was to allow for a much larger loss, cut down on the amount of trading, and in the end have a larger profit. This is the option we took.

So why would remaining in buy mode when the market is declining be a better solution? The stock market will often have corrections during a larger trend, whether that trend is up or down. When employing option one, most often we would be selling right near the bottom of the correction. In the most recent case this did not happen, but often it does. It is not because we have not chosen better indicators for our option one, it is because of the nature of how the stock market works. In October of 2005, using option one we would have changed our signal to sell right near the bottom of the decline. By the time we switched it back to buy, a 4%-5% loss would have been the result. Using option one, the current market is far from being a buy. By the time we would trigger a buy using this option, most likely the market would be significantly higher and our gain from the May 15^th sell signal would be minimal. In the end, large drawdowns are inevitable and we believe we have chosen the best way to deal with them. Option one leads to more trades, which leads to higher commission costs and more frustration, and is less profitable in the end.

Many of our competitors have had a handful of trades during this decline. These trend followers are selling at bottoms and buying the tops. Trading so often leads to higher commission costs and is just unnecessary. You want to eliminate these kinds of swings when designing a system. We have also read on different sites that people are considering altering their system due to the current fall! This is a decision that will most likely lead to disaster and the end of their timing system(s). The fact is, this market has been very difficult to time and if you are following a timer who believes the answer is to change their model because of it (after the fact), then we strongly recommend that you stop following this person’s advice. It is hard to accept that you will inevitably have drawdowns in any system you choose, but this is a fact in a game where you are trying to predict the future by looking at what has already happened in the past.

We knew that the current situation was inevitable; we just hoped that it happened later rather than sooner. Regardless, we follow the plan of action that we put together when we designed our systems. We have a very late trend following aspect in both of our systems and if the market declines enough, we will have to sell. If this happens, the sell will most likely be around a break of the 2000 area in the NASDAQ. If the stock market is going to hold its lows, it needs to hold them now. This week is very important and could be a pivotal week. Perhaps the bottom will occur (or already did occur on Friday) or the downtrend will continue and we will end up having to have a late sell signal.

It is important to note that we are surprised that this decline happened here, but we are not surprised that this sort of decline could happen and surprise us. Meaning, we knew that there would be significant losses in every timing model we could possibly design and to avoid multiple trades that ended in higher losses over time, we would occasionally have to sit through a large correction and maybe even flip the signal if the correction got large enough. 

This will most likely be an important week. Our eyes are on the 2000 level on the NASDAQ Composite, even though our system, as always, will tell us if we need to be late-defensive and sell.