Can I get a few thoughts on these backtest results? - page 3

New comment
 

No, it's the same Idea from my SDP-Tread. Close 1/2 the size and let the rest ride. Allow me to be critical for a moment. The problem with trend systems is Pace. This will run havoc on most people physic. I mean you traded for a whole year and still where you started, taking small losses over and over.

I don't even know why people create 1000s of different indicators. 99% of these give the same info. Example, if I placed Stochastic on a chart and played with the period enough, I could create nearly the same effect as your Blue line. Whose to say your blue line is superior. So, I agree with you, use what ever makes you feel better :). Reason why I made that statement and I almost forgot, is because most trending systems would show similar results, your system may show similar results to the system presented in my link as you begin to go back in time.

Since, you're a trend follower, I'll recommend Pyramiding the lots as position moves in your favor. (experiment with this if somehow you don't like the 10 years results). Trend-traders really have to take maximum advantage of the trend. This is your bread and butter. Offense is your greatest defense type thing. Example: where you closed that second order, because blue & green were still above 0, that may have been a good point to stack ... while moving the trailing stop of course :).

Anyway still like your system and look forward to the 10 year results if you choose to provide.

 
Actually, now that you mention it, the other day I tested the same 6 month period and instead of using 2% of my balance for each trade, I used 5%, and the results shot up to >$30k profits. So I can agree with that, the trend is the bread and butter and I should utilize that to my advantage. I just need to find that perfect spot that I'm happy with risking. :-) I'm working on the 10year backtest, will post shortly when it finishes.
 

Well that's just sad. :-(

jan 1, 2001 - Present

 
Not really lol... Ask yourself if you can live with those results for 10 years. I know I cant. Time for more experiments. However the fact that it's working in recent times should be an encouragement to use Now. But you knowing this will make it difficult as you'll not be able to trade it further than you can throw it.
 
So I just opened the chart and the first date it has is Dec 12, 2005. I'm guessing my broker doesn't provide anything before then. So that was only 5 1/2 years or so.
 

Download 10 years of data here. Use period converter script to turn M1 into all periods. Things to keep in mind: it's not your broker's data. But your broker's data is not Perfect to real time either because it's only M1. Worse case scenario, it'll tell you how your system performs on other brokers data. Forex is not an exact-science IMO. It take only 1-pip difference to make test on one broker look more attractive than the other. Your system should be built to over-come any and produce similar results.

 
wow, it seems like there are infinite variables in dealing with forex. :-)
 

Made a small tweak to my open position. Tightened up the restrictions a little. It takes way less trades, but yeah... check it out. :-)

Again, this is over the last 6 months, where I get the best results already.

 
actually, ignore that. toward the end my lots calculator got all messed up and started giving ordersend errors because it couldn't divide the lot into 2 evenly. lol
 

Not enough trades IMO. You're now walking the Curve-Fit route and on small number of trades at that. Those results are no-where close to what you can expect. Set realistic expectations.

Why not just forward test it? It might make you excited for a week or so (if even that). 8)

1234
New comment