According to Paul Desmond of Lowry’s Report : We found that almost all periods of significant market decline in the past 65 years have contained at least one, and usually more than one, day of panic selling on which Downside Volume equals 90% or more of the total of Upside Volume plus Downside Volume, AND on which Points Lost equals 90% or more of the total of Points Gained plus Point Lost. I do not have the time to put together a program to calculate the Points Gained portion of the above equation.  A task for the near future. However, below is a chart of the S&P 500 marked with 90% Down Days, as you can see there have been quite a few.

Market declines containing two or more 90% Downside Days often generate a series of 90% Downside Days, often spread apart by as much as 30 trading days. Therefore, it should not be assumed that an investor can “ride out” such a decline.

Bank of America (BAC) is in talks with the Fed looking for more TARP money.  JP Morgan (JPM) announced earnings declined 76% this morning although that was a bit better than expected. And of course what everyone knows by now Steve Jobs of Apple Computer (AAPL) will be taking a medical leave of absence till June 30. Steve Jobs is a man I think almost everyone admires and I wish him all the luck in the world and a speedy recovery.

On the economic front this morning: the PPI was -1.9% vs. expectations of: -2.0% and ex food & energy +0.2% vs. expectations of : +0.1%.  Empire State Manufacturing Survey came in at -22.2 vs. expectations of: -25.0. Rounding out the data Jobless Claims came it at 524K vs. expectations of: 500K.  Then at 10 AM the Philadelphia Fed survey Consensus Forecast for January 09: -35.0.

Overseas markets are a bit lower: Nikkei 225 -4.92%, Hang Seng -3.37%, Shanghai Composite -0.45%, DAX -0.54%, and the FTSE 100 -0.52%.

Shortly after the release of the economic data the futures: Dow -65, S&P -6.50, Naz -14, Oil -$0.90, and Gold +$3.00.

Yesterday I back tested a simple moving average crossover system against the QQQQ, since the inception of the QQQQ approximately 18 years ago.  If you downloaded the summary and trade list and looked them over you noticed the overall results were mediocre.  It’s not the purpose of these posts to give you the “Holy Grail” trading system against the Q’s or any other vehicle. I’m merely starting from a conceptual point, back testing that concept, and moving forward with an analysis of this and in the future other trading methods. 

Yesterdays test was an always in the market test, i.e. you are always in, either long or short.  Having spent many years in the brokerage business I’m well aware of the public at large unwillingness to sell short.  Therefore, I’ll perform the same test as yesterday but from the long side only, i.e. you either own QQQQ or you’re in cash earning interest. Interest at money market rates are not considered in the comparison results:

As you can see the long only trading outperformed the long and short trading by almost 3 to 1 on a risk adjusted basis. The maximum drawdown is also quite a bit less with the long only trading -46% versus -53%. Obviously therefore we should trade this system long only; right?  Not so fast. Well the next step is to look over time frames and see if something else is at work, and the answer to that is yes:

In other words, after the 2002 bottom, trading conditions changed.  No more extraordinary trends after the bubble burst.  Therefore we run the same back test once again but only for the shaded area above.  Results: Long + short trades return -1.57%, long trades +3.35%, short trades -9.02%. 

What you must watch for with any methodology is a change in market conditions, in this case there are no more extraordinary trends to show a result that in essence is no longer reasonable to expect, at least from the long side only. Will other moving average crossovers test better? I’ll get into that perhaps later this week or next.  There is a lot more to developing a reasonable system than just a computer back test.