Rising unemployment and falling industrial production bode ill for the economic recovery and with it corporate earnings. Investors have been nervous for some weeks that the market's gains have outstripped earnings potential and in the last two weeks that nervousness graduated to fear. That fear has started to have a tangible impact on stock prices as we shall see.

As usual we consider the NASDAQ Composite because it is from that exchange that the most profitable breakouts arise. The one year weekly chart shows that we just endured the first back-to-back weekly loss since the recovery began on March 9. This was also the biggest weekly loss both in points and percentage since March 9. The index is testing support at the 10 week (50 day) moving average level which it hasn't done since the first week of July. At that time, the index recovered strongly after a double test of support, in part due to the 'cash-for-clunkers' program which came on-stream in July. With no such stimulus to lift earnings prospects now, manufacturing still contracting and unemployment rising, we could well see the 50 dma support level breached decisively in the next few days.

If the 50 dma support level is breached then the next significant support level is at 1880, an 11.6% drop from Friday's close. A breach of support would be a signal to consider shorting some of the stocks on our short watchlists. On Friday we issued ten confirmed breakdown alerts, a sure signal that there are now shorting opportunities with more to come.

Not really a top tip (yet) but interesting nonetheless.

A subscriber has hypothesized that breakouts are more successful when the volume on the day after breakout exceeds the breakout day volume. He also hypothesized that the failure rate is lower in that case.

A preliminary look at our data suggests that the gain after breakout is indeed superior in this case but that the failure rate is not improved. This deserves more study over a longer time period (we can go back to April, 2003) to confirm the result. We also need to determine if the improvement in return offsets the (possibly) higher price that would be paid by waiting for the second day's volume. We will study this next week.

Do you have a system that you believe works but needs more rigorous testing and analysis? If so let us know and we will backtest it against our historical data and publish the results.