Tuesday, August 25, 2009

Qs Happy Trails Revisited

This is a cleaned up version of an earlier series of systems I termed Happy Trails. Run through the posts on the IWM and the Qs to view the logic and various stops applied, including the effects of pyramiding.
For this version I've changed the entry/exit signals to fire "Next Bar at Market" as opposed the previous "at close". Perhaps surprisingly, this slight change produces a nice little net gain. I've also reduced the inputs from 3 in the original Qs version back down to 2 as in the IWM system.
This system trades a lot. . .125 trades in 5 years, or a trade every 2 weeks, but as you can see from the performance report the actual average holding time is only 2-3 days.
Reviewing other performance metrics I've discussed previously, the max consecutive losers is rather astounding, as is the max consecutive winners. The max intraday drawdown is also impressively low.
Now, before you start deploying all your capital to this little gem there are a couple things to keep in mind.
First of all there are 2 stops conditions in place. One is a breakeven floor stop. . .set at $4, and the other is a stop loss (both long and short sides) set at $90. All trades are 100 share positions and they are tracked frictionlessly....meaning no commissions or slippage applied.
The net result of the $4 breakeven floor stop is that we get a lot of trades that go nowhere. . . as reflected in the flat period of the equity curve shown below. Nevertheless, there are a number of simple conditions that can easily tone down the trade frequency and reduce the equity chatter.
The trade distribution above shows the predominance of winning trades.
And, for the backtesting critics . . . I've taken the system time parameters and moved them ahead 11 months to the current (shown below).
Performance levels continue along the same track and the equity curve has shown a marked jump (not shown).
Thanks to Vimal for questioning me about Happy Trails and forcing me to investigate its nuances in more detail.
TS2000i code is shown below. The breakeven floor stop and stop loss conditions have to be added independently to the system parameters.


Toptick said...

For performance reporting, new TradeStation treats trades as either win, lose, or breakeven, so the percent profitable drops below 20% because so many trades are stopped at breakeven. Commission and slippage would have an impact.

Interesting. Thanks!

bzbtrader said...

Glad you're still me. As always, these studies have to be considered in context. . as part of the larger fabric of my basket of systems and my stated intent to search out non-correlated technical confirmation signals. In that respect I believe Happy Trails is a valuable asset. I continue to explore nuances of this oscillator and will post updates as time permits.
Thanks for running the Trail through the TopTicker filter.
As I mentioned in the post and you reiterate, commissions can significantly eat into the net gains, but if you trade on a per share basis the equity curve and the net returns still look good to me.