First of all, let me offer my sincere thanks to Brandon for keeping the blog up-to-date with Power Dip signals while I was on vacation. The system will be closing out a few trades tomorrow morning and will be replacing those positions with new stocks. Look for the post in the morning.
I have updated the spreadsheet and I am happy to note that the system’s real-time results (with 70 trades logged) are beginning to come close to our historical backtested results. For example, the average trade is now approaching 0.70%, which is pleasing to see since it has been hanging around 0.40% for a month or more. I expect that it will eventually approach 0.90%-1.0%, which is what we expect from backtesting.
The win% is very close to the historical backtested average of ~68%.
Keep in mind that I have been diligent in accounting for commissions, and whenever possible, I have included the prices of actual fills, to account for slippage. What you are seeing are actual, replicable, real-time results of a dip-buying system.
Please note…
The system has been keeping 20% of available capital in cash, and has been using percent-risk position sizing, holding risk per trade at 1% of total capital. We can juice returns by allowing the system to use all available cash and by using 1.5% or even 2% risk per trade.
Over the next week, I will be posting the results of using all of the available capital with more at risk per trade.
If there are any questions about how percent-risk position sizing works, or how it can be manipulated to affect returns, please do not hesitate to ask.
Hey Wood. Good to have you back. My similarly developed system is showing similar results (66.3% wins), 1.92% gain average per trade. I use a 1.5% risk per trade and then occasionally bump it up to 2% for trades I deem higher probability, based on my research. For me, the hardest part is keeping my money at work. I use a position size based on volatility, which allows me to carry about 15 positions at any one time (with all the cash at work). While this has improved my risk profile, it is at the expense of commissions. I find your 10% flat stop intriguing in that it would mitigate some of that issue (relatively large number of positions) but would make it more difficult to scale into positions (I typically take on no more than 3 new positions per day). Anyway, nice work with the system you’ve developed. The Fly frowns on mechanical systems; but The Fly is gay and doesn’t understand their utility. Indeud…
Nice results Junk!
I’m guessing if I ramp up our percent risk we will get average trades that are close to what you are getting.
I like volatility based position sizing, but I understand exactly what you mean as it can have you in some really small positions sometimes.
I need to test scaling into the number of positions on AmiBroker. I too agree that intuitively it seems better to scale into these mean reversion setups just in case a load of them never revert 😉