Great EA in backtest! - page 135
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
to report. My fxdd account seems to work fine. I have not seen the spread go above 3 pips. That's a refreshing change from IBFX which often exceeded 6 pip spreads on the euro totally destroying the CT program.
The version I am currently forward testing is 1.95 SR Euro-variable SB. As such I have the spread filter and an minimum ATR filter in place in that version. For the past three days the market has not given a ATR in excess of .002 which is the minimum which that filter allows to trade therefore it has not opened any positions for the past three days.
I got quite desperate to make something work with this before I finally concluded that ibfx had to go. It is possible that my filtering is more restrictive than it has to be. I am looking at back tests now from version 193b and this version and wondering if it's an advantage or not to lighten up on the filtering in favor of getting more orders. I have not reached any conclusion about this yet nor have I really conclusively decided which months are best or worst. I would like to see Brazilian traders data that leads him to conclude that January's are good. how strong of a monthly pattern do you really see and over how many years did you test to arrive at this conclusion? Can the results be explained by other market conditions besides the month?
the primary difference between 193b and the 195 version I'm using, besides the spread filter itself is the ATR filter. That is what is primarily responsible for keeping me out of the market for the last three days not the spread. I am not convinced that a setting of 0.002 is optimized based on the comparisions with 193b which has no ATR minimum filter at all and shows more reward for it. Perhaps I'm being overly restrictive now that I'm with fxdd??
to report. My fxdd account seems to work fine. I have not seen the spread go above 3 pips. That's a refreshing change from IBFX which often exceeded 6 pip spreads on the euro totally destroying the CT program.
The version I am currently forward testing is 1.95 SR Euro-variable SB. As such I have the spread filter and an minimum ATR filter in place in that version. For the past three days the market has not given a ATR in excess of .002 which is the minimum which that filter allows to trade therefore it has not opened any positions for the past three days.
I got quite desperate to make something work with this before I finally concluded that ibfx had to go. It is possible that my filtering is more restrictive than it has to be. I am looking at back tests now from version 193b and this version and wondering if it's an advantage or not to lighten up on the filtering in favor of getting more orders. I have not reached any conclusion about this yet nor have I really conclusively decided which months are best or worst. I would like to see Brazilian traders data that leads him to conclude that January's are good. how strong of a monthly pattern do you really see and over how many years did you test to arrive at this conclusion? Can the results be explained by other market conditions besides the month?
the primary difference between 193b and the 195 version I'm using, besides the spread filter itself is the ATR filter. That is what is primarily responsible for keeping me out of the market for the last three days not the spread. I am not convinced that a setting of 0.002 is optimized based on the comparisions with 193b which has no ATR minimum filter at all and shows more reward for it. Perhaps I'm being overly restrictive now that I'm with fxdd??I made a 3 years test with CT 1.93b and Templar did one since 1999, our data feed is from MT4 servers, since we downloaded it from the platform...
I commented earlier why I think CTs performance is mediocre at the end of the year: because of lots of up trends as well as the majority of CT's orders that is sells.
Im trading live with CT now on NorthFinance. The 2 pip spread is a blessing and is essential to get CT to work properly. I shouldnt have traded this week in thin conditions but i had to make some comparative tests to demo/backtesting. I started it boxing day and the second trade hit my stop loss which is a painfull start but its was verified on backtest as real loss despite hit was within 2 pips of stop position. The rest of the week has been ALL winners small 10 pips here and there pip perfect to backtest over same data. but i closed it few hours early as i did not want it open this weekend which can gap over 100 pips in theory the highest risk weekend of the year. That proved to be another mistake as CT would have go on to make 2 more winners in the closing hours but still i couldnt risk been stuck in a bad trade over New Year holiday.
So good news and bad news really it does mostly what its supposed to do but condition not right to make good money. Again backtest show that it can take till Aprill before this thing really finds its grove and flies and i think if i get to 1st week of Feb at least as break even it has a very real chance of working. I can also point out the last 2 months have been the worst for CT compared to any month the past 3 years! I hope its just a dip and recovers next year but i blame that on all time low ATR.
One final point after exhaustive testing i now got CT to work with NO filters, no cci adx nothing in that section just some good tweaks to the CT logic. This is exactly what i needed as i believe anything added here gives bad bias to decisions.
Anway the quicker you get away from IBFX and FXDD the better. Albeit IBFX are scammers and i hope you realise all trades are cleared through another broker and IBFX is just an IB for FOREX LIQUIDITY who will put you on manual if you make a profit.
Im trading live with CT now on NorthFinance. The 2 pip spread is a blessing and is essential to get CT to work properly. I shouldnt have traded this week in thin conditions but i had to make some comparative tests to demo/backtesting. I started it boxing day and the second trade hit my stop loss which is a painfull start but its was verified on backtest as real loss despite hit was within 2 pips of stop position. The rest of the week has been ALL winners small 10 pips here and there pip perfect to backtest over same data. but i closed it few hours early as i did not want it open this weekend which can gap over 100 pips in theory the highest risk weekend of the year. That proved to be another mistake as CT would have go on to make 2 more winners in the closing hours but still i couldnt risk been stuck in a bad trade over New Year holiday.
So good news and bad news really it does mostly what its supposed to do but condition not right to make good money. Again backtest show that it can take till Aprill before this thing really finds its grove and flies and i think if i get to 1st week of Feb at least as break even it has a very real chance of working. I can also point out the last 2 months have been the worst for CT compared to any month the past 3 years! I hope its just a dip and recovers next year but i blame that on all time low ATR.
One final point after exhaustive testing i now got CT to work with NO filters, no cci adx nothing in that section just some good tweaks to the CT logic. This is exactly what i needed as i believe anything added here gives bad bias to decisions.
Anway the quicker you get away from IBFX and FXDD the better. Albeit IBFX are scammers and i hope you realise all trades are cleared through another broker and IBFX is just an IB for FOREX LIQUIDITY who will put you on manual if you make a profit.Good... Which version are you using? Did yourself do the tweaks on its logics?
Yeah, FXDD as David said (he's live trading with them with CT) has fluctuation on their spread, but it happens for seconds and never more than that, only on REALLY insane price variation... it never happened when CT was placing any order, so he said there is nothing to worry about with FXDD. I would prefer FXDD because I don't trust NF. I don't know... their website template is scary... seems too much a Scam... and I have heard bad things about them.
Bolt, I would like to know what tweaks you have made to the logic. I have spent considerable time studying the logic and code of this EA. I believe I have the optimal settings for the 193b version but I'm not so arrogant as to be closed to the possibility that there isn't a better revision. So what I really have is the best settings and configuration that I am aware of. If you have something better by all means please share it.
From what you said it sounds to me like you don't have an in depth understanding of the code but I would love to have you prove me wrong ya know?
I am glad to hear that it performs in a live account like it demonstrates on the backtest. I too am leary of NF. At this point I'm not really comfortable with them. I'm not really comfortable in my financial situation and that's the core problem. If I had multiple accounts spread around and felt like if I lost one of them it wouldn't hurt me it would be different. But that's not the case. I'm just starting to build a nest egg and any loss is painful. I'm just very protective and know that I'm vunerable.
I have watched the fxdd account for the past week and I have seen the spread vary one pip but no more. I am currently allowing the 193b to trade live on my account with risk=0.1 as such I expect my $760 account to open it's next position at around .06 lots which will be moving .60 cents/pip.
Since I know how fond we all are of posting our backtest reports on this thread, here's mine now that I have 7 years of data. If I were gutsy enough to allow it to trade with risk = 1.5 this is what it would look like...
thing is that in short order this backtest reaches the maxlot=10 on the mini account and at that point the risk settings become a moot issue. What you call 'flying' I guess. I think this EA will fly once you reach a point in your account where at maxlots you will not suffer an unrecoverable drawdown in your account. I'm not certain where that point is exactly. I still have to dig into this test and see just how long the worst drawdown happened time wise. and then see if that was during a maxlot=10 how much would that represent in dollars?
Anyway, I'm following David's lead on this with the brokers since he seems to have actually made the best returns of anyone on the thread to date that I know of. I may still trade at a higher risk level than he does once I am satisfied myself that it's performing according to the backtest model. Time and trades will tell.
my goal now really is just to see it make more than it loses. I never could get that out of ibfx.
]Anway the quicker you get away from IBFX and FXDD the better. Albeit IBFX are scammers and i hope you realise all trades are cleared through another broker and IBFX is just an IB for FOREX LIQUIDITY who will put you on manual if you make a profit.
NorthFinance is no better. Is this the same bolt from the forexnews forum?
I think I am being overanxious I can prove the pattern with smaller lots just as well as with larger lots. Once the pattern has shown to manifest from the live account THEN I will increase the risk...Until then..I will run with risk=.02 which will give me lot sizes of .01
It shouldn't take long to know.
I think I am being overanxious I can prove the pattern with smaller lots just as well as with larger lots. Once the pattern has shown to manifest from the live account THEN I will increase the risk...Until then..I will run with risk=.02 which will give me lot sizes of .01 It shouldn't take long to know.
yeah, that is a good strategy... I wish you success!!
Bolt, I would like to know what tweaks you have made to the logic. I have spent considerable time studying the logic and code of this EA. I believe I have the optimal settings for the 193b version but I'm not so arrogant as to be closed to the possibility that there isn't a better revision. So what I really have is the best settings and configuration that I am aware of. If you have something better by all means please share it.
From what you said it sounds to me like you don't have an in depth understanding of the code but I would love to have you prove me wrong ya know?
I am glad to hear that it performs in a live account like it demonstrates on the backtest. I too am leary of NF. At this point I'm not really comfortable with them. I'm not really comfortable in my financial situation and that's the core problem. If I had multiple accounts spread around and felt like if I lost one of them it wouldn't hurt me it would be different. But that's not the case. I'm just starting to build a nest egg and any loss is painful. I'm just very protective and know that I'm vunerable.
I have watched the fxdd account for the past week and I have seen the spread vary one pip but no more. I am currently allowing the 193b to trade live on my account with risk=0.1 as such I expect my $760 account to open it's next position at around .06 lots which will be moving .60 cents/pip.
Since I know how fond we all are of posting our backtest reports on this thread, here's mine now that I have 7 years of data. If I were gutsy enough to allow it to trade with risk = 1.5 this is what it would look like...
thing is that in short order this backtest reaches the maxlot=10 on the mini account and at that point the risk settings become a moot issue. What you call 'flying' I guess. I think this EA will fly once you reach a point in your account where at maxlots you will not suffer an unrecoverable drawdown in your account. I'm not certain where that point is exactly. I still have to dig into this test and see just how long the worst drawdown happened time wise. and then see if that was during a maxlot=10 how much would that represent in dollars?
Anyway, I'm following David's lead on this with the brokers since he seems to have actually made the best returns of anyone on the thread to date that I know of. I may still trade at a higher risk level than he does once I am satisfied myself that it's performing according to the backtest model. Time and trades will tell.
my goal now really is just to see it make more than it loses. I never could get that out of ibfx.instead of backtesting 7 years, can you backtest month of may until december individually???
I think I am being overanxious I can prove the pattern with smaller lots just as well as with larger lots. Once the pattern has shown to manifest from the live account THEN I will increase the risk...Until then..I will run with risk=.02 which will give me lot sizes of .01 It shouldn't take long to know.
have you forward tested cyberia? how long has it been?