Hi alphadude,
when i think of someone who has experience in automatic systems and stuff you come to my mind, and i was wondering if you have seen something about trading the equity (in the way im going to tell you). From some of your posts i read i think you're trying to get stable growing equity curves without overfitting. But imagine that you backtest a (doable) strategy and you get a equity with perfect sinusoidal shape that has had many oscillations to make it better; since that strategy doesnt fit your criteria (what i think is your criteria, because objectively it has 0 net profit) you discard it, but from my intuition it could have huge potential.
I dont know if its even possible to find such kind of strategy, it wouldnt have to make a 'perfect' wave equity, even it could have a small negative slope, but you get it... (finding a not-overfitted steady growing strategy is also pretty unusual anyways)
I know that even we find it, it could just break and become unstable in the next cycle, but i'd still would give it a go.
By filtering that strategies and trying to make some sense of why they work that way, you could have a bigger portofolio.
Also it doesnt need to be sinusoidal, there are many kinds of exploitable shapes.
Hey all, I recreated the raw strategy alphadude has posted. To this day there is still edge. The edge has been proven and is imho enforced by fundamentals. I will be developing this strategy further. Thanks for sharing alphadude. If anyone wants to share ideas or theories i'd be happy to discuss those. Take care!
Hey all, I recreated the raw strategy alphadude has posted. To this day there is still edge. The edge has been proven and is imho enforced by fundamentals. I will be developing this strategy further. Thanks for sharing alphadude. If anyone wants to share ideas or theories i'd be happy to discuss those. Take care! {image}
Ignored
I improved the strategy as best as I could.
No lagging indicators - these all gave bad results
Removed the closing at end of day and instead exit on reverse signal - this improved results by a ton
Add to winning positions with a bigger lot size - this improved results by a ton
When added to a winner protect those positions
News filter
Can increase initial range time from 02:00 to 04:00. I decided to leave it as standard
Tested on multiple pairs. - USDJPY works best
By doing this, we can increase the profits by at least 50% over the full back test
When adding automated position sizing it will grow even more.
I am going to live test this strategy on the coming monday.
