Let's add the following stop-loss to the most recent combination of days-of-the-week indicators and filters.  If long, subtract 66% of the average three-day range from your entry price (the open).  If short, add 66% of the average three-day range to the entry.  If you survive the stops, as always, you're out on the close.  The results follow. 

Pretty exciting, if you ask me.  We're not playing fast and loose with the stopâ€"it's applied the same way across the board.  Not only that, but it's dynamic as opposed to inflexible.  It's much easier for straight money management stops to go bad than ones that increase or decrease as market conditions change.  You could probably get away with a three- or four-point stop in the last couple years, but they would prove ridiculously tight in the bubble environment of the late 90's.  When you key a stop to a market characteristic such as range sizes, you don't have to guess whether they're going to become too narrow or wide as the markets drift.

Let's look at one more chart.  The days-of-the-week driver makes sense in a stock-related, widespread-investor psychology environment.  That's why we don't expect to ever use it in the yen or bonds.  However, just as a point of interest, look at how this exact same idea holds up in the other financials we've tracked.

Again, not a suggestion that there are more markets to exploitâ€"but more corroboration is always more fortifying than less.

Let's give this system a nameâ€"no doubt we'll be referring to it in days to come.  Let's call it DYNAMIC DAILY ONE.

Art Collins is the author of Market Beaters, a collection of interviews with renowned mechanical traders. He is currently working on a second volume. E-mail Art at artcollins@ameritech.net.