The Trading Digest

The massive runup from the extreme oversold lows of Mid-August gave me an idea to test. Let’s test a rate of change that just occurred from the lows on the Nasdaq 100 37 days ago. As of Fridays close we were greater than 12% higher than price 37 days ago. A big big move.

So if we screen for that condition and enter the next day at the open. (noted in blue dot)

And code an exit logic that looks like this:

We exit no matter what after 22 days (about 1 month) at the following open

or

We hit a 10% profit target

We do this on the Nasdaq 100 Index going back to Feb 1988 (beginning of data)

This condition triggered on 62 occasions - 69.35% were profitable - with an average profit of 2.65% - the dollar profits and drawdown are based on $15,000

Thats massive. As a matter of fact the average return over any 22 day period would be +1.22% yet we are seeing +2.65% average return.

The 10% profit target probably was only hit once or twice right? Wrong. Out of the 62 trades 16 hit the 10% target before the 22 day timeout exit. That is about 25% of the time. I put a link below where you can download a PDF showing all the trades.

Here is a PDF with all those trades listed

I am not here to expouse a particular bullish or bearish view. I can only show you what I am seeing in this test. I have no bias and I will not make any particular trade based upon this test. This data is very similar to a test I ran on the old blog that everyone dismissed at the time (here and here). Also some of the newer readers may benefit from a post on the old blog that talked about my portfiolio construction. That has changed only slightly from when I wrote that and has done very well for me. But that is another post for another day.

Have a Great Day!

Dave Johnson

2 Responses to “An Interesting Backtest I Ran”

1. Damian Says:
   October 9th, 2007 at 10:06 am

   Dave - what’s the condition? ROC 12%? From the trades it looks like > 12%….

2. Dave Says:
   October 9th, 2007 at 10:11 am

   correct a close >12

Leave a Reply