"...work through each day of past history that you want to check. It's pretty easy to find the trades. " I started back in July 2002 and found 86 trades. 43 winners, 3 that I counted at breakeven (because they went to the close at about entry price) and 40 losers. That would be a 52% win/loss ratio. Your profit factor could well be the 3:1 that you stated earlier as I did not calculate dollars on each trade, just win or lose. For me, there were just too many days where there were no trades. It looks like it would lend itself to automation fairly well though. Best wishes - rcm.
Jeepers I've only looked back as far as June of this year so far. Found a trade almost every day with a much higher win to loss ratio. There would have to be a radical difference for the number of trades over time to drop so much in the previous year. Honestly no offense intended, but are you sure that you're looking for the setups properly? Really seems off to me. I'll study backtested results over the past year or so over this weekend and post my findings if anything relevant comes of it. I agree that it looks like it can be automated. I have some divergence signals (Stochastics) automated already in TS2000i, so this should be doable.
MarkB, I'd be interested in backtesting this method in Wealth-Lab but I don't have any code that automatically detects divergence. If it's possible, would you mind emailing me your TS code (in plain text format). I could then translate it into WL for use with the Chailkin oscillator and then backtest it. If it's too much to ask, that's fine; I'll try write a divergence function from scratch myself. Richard
I'm not an Easylanguage expert. Getting better at it, but not there yet! Actually the divergence code was written for me by someone else. I can use it with the stochastics, but the file itself is protected (tied into my TS customer ID too). Which, by the way, was NOT according to our deal. I'd love to break that protection if anyone knows how, since the guy didn't honor our contract and has vanished. Meanwhile though, I know it can be done, so I'll find a way to code it.
"Jeepers I've only looked back as far as June of this year so far. Found a trade almost every day with a much higher win to loss ratio. There would have to be a radical difference for the number of trades over time to drop so much in the previous year. Honestly no offense intended, but are you sure that you're looking for the setups properly? Really seems off to me." Yep, it definitely got on a roll since June. And I may well be not seeing the trades the same as noPM - but I do have the exact same setup with the big SP etc. The trades are actually pretty easy to see...rcm
Interesting RCM..... I'm looking forward to checking things out back in time then. Comparing notes is interesting! no_pm, for how long have you been applying this technique as is?
Thanks, I'll have a look at that. Took a peek and I'm not sure that it applies to this case. Better that I know that it's doable, so I can learn to do it myself instead of trying to pierce this guy's code. Let's not get off topic here though. I think we all owe a round of applause to no_pm for sharing his method with us.
no_pm, This is an excellent thread. I especially like that you have described your method in a very mechanical easy to understand way. Thank you for sharing it.
Well I hope I can analyze some historical data this weekend. If anyone's counting on me..... well, my mother managed to sit on her glasses tonight. So tomorrow's gonna be a scramble for new frames at least! Perhaps new lenses. There goes the profit for the week, hehe.