You are using an out of date browser. It may not display this or other websites correctly.
You should upgrade or use an alternative browser.
You should upgrade or use an alternative browser.
Just show me one mechanical technical analysis that has been profitably backtested?
- Thread starter crgarcia
- Start date
Ah what the heck, you can all have this one for free.
LONG
When the LOD precedes the HOD, buy the LOD and cover at HOD
SHORT
When HOD precedes LOD, sell HOD and cover at LOD
A couple of timing issues relating to CME application.
This strategy will leave you flat on saturday and sunday.
See nothing is perfect.
regards
f9
LONG
When the LOD precedes the HOD, buy the LOD and cover at HOD
SHORT
When HOD precedes LOD, sell HOD and cover at LOD
A couple of timing issues relating to CME application.
This strategy will leave you flat on saturday and sunday.
See nothing is perfect.
regards
f9
Quote from fearless9:
Ah what the heck, you can all have this one for free.
LONG
When the LOD precedes the HOD, buy the LOD and cover at HOD
SHORT
When HOD precedes LOD, sell HOD and cover at LOD
A couple of timing issues relating to CME application.
This strategy will leave you flat on saturday and sunday.
See nothing is perfect.
regards
f9
LOL NICE!
Please be avare that us-stocks data is practically worthless for backteting because of false ticks.
I accept stock backtest results only if they use market orders in and out.
Limit or stop orders give dramatically different results compared to real trading.
The ones buying at 10% below yesterdays close and selling next day have profit factors of 2 in bt and in real trading lose.
My current system with market orders since 2000:
- 710 trades
- Profit factor: 2.68
- max. DD 14%
- annualised gain: 63%
- wl score: 192
- long only
Ultra simple and not buying crashing stocks like most dippers do.
I accept stock backtest results only if they use market orders in and out.
Limit or stop orders give dramatically different results compared to real trading.
The ones buying at 10% below yesterdays close and selling next day have profit factors of 2 in bt and in real trading lose.
My current system with market orders since 2000:
- 710 trades
- Profit factor: 2.68
- max. DD 14%
- annualised gain: 63%
- wl score: 192
- long only
Ultra simple and not buying crashing stocks like most dippers do.
Quote from Glowy:
lets say you are good, real good
why would you be living in Argentina
serious question ?
I don't know if you traveled a bit around but much much better places exist
We have been living between Arg and Brasil and traveling for several months each year (apart from this one)
We are off this weekend to Panama, Puerto Vallarta for Xmas with LA amigos, Miami and back to Salvador for carnival, London and back in BA around May and back to Brasil.
We like the latino lifestyle and the time zone is good.
Where do you think is a better place to live.
regards
f9
Quote from Fishbird:
Please be avare that us-stocks data is practically worthless for backteting because of false ticks.
I accept stock backtest results only if they use market orders in and out.
Limit or stop orders give dramatically different results compared to real trading.
The ones buying at 10% below yesterdays close and selling next day have profit factors of 2 in bt and in real trading lose.
My current system with market orders since 2000:
- 710 trades
- Profit factor: 2.68
- max. DD 14%
- annualised gain: 63%
- wl score: 192
- long only
Ultra simple and not buying crashing stocks like most dippers do.
very true. real life and "academics" of back testing are very different. even results from going live on paper account(like one IB provide) and real fills can be very different. specially on stocks. picture from WL, i posted above, shows great performance,but in real life it's nearly impossible to get into good trade and of course the losers are all yours.
Quote from Glowy:
my system even outperforms yours
yours took 8 years if one trade a day was placed to go to 1 mil
too long because you won't allow for larger DD
You're out in the public domain, and we are referencing publicly available scripts. As far as I'm concerned, you're just blowing steam. I can code a perfect script with 100% accuracy and post the backtest, but how is anyone going to know if it's real and doesn't peek?
You're just bullshiting without seeing the actual program.