ForexMorningTrade System

[redlion]

Since 1 January. 9 trades opened at 06.30 UK time.
5 wins 4 losses @tp28/40sl no break even resets or trailing SL. down 20 pips

latest 100 signalled trades, extending back to mid July 2010
At 28TP +760 pips 70 wins
At 35TP +425 pips 59 wins
At 40TP +400 pips 55 wins

Hi, I'm litte bit confused about trading time you published. According to TMT manual, page 5, 6 all trade are checked(executed) on 5.30 GMT. Alpari UK ( im also using them ) operating GMT +1hour. And at the moment in UK ( UK time ) =GMT time. So wiseambitions , your trades ( If you operate from Alpari UK)should be executed on 5.30GMT( which at the moment is UK time) or 6.30 am Alpari UK. Do not be offended, I just want explanation cos im confused.
Cheers

 

[billcurry]

The whole point of my exercise was to show how important it is to do back testing on any EA you are serious about, with good data. Good data, that is very important!

Where/how do you get your data for backtesting? I tried getting tick data from Dukascopy per this tutorial but I was getting some strange errors from FMT. It would run fine and it would finish with 99% modeling quality, but there would be no trades. The journal showed that it tried to trade but I kept getting "ModifyOrder error 4051" because for some reason it was setting the lot size to 0. I checked my settings and everything looked good, so I'm not sure what is causing this error. I'm guessing there may be something wrong with the tick data I downloaded/converted.

 

[thruster9121]

Have a look at this

Good tutorial on getting a 90% quality, it might help.

 

[thruster9121]

Also Alpari 1m data gets good tick acuracy...http://www.earnforex.com/blog/2008/01/quality-metatrader-historical-data/

 

[RedGuerrilla]

Hammy,

Based on Red Geurilla and Your testing, would it be safe to say that a start time of 6:15 and 6:30 would be the best times to trade FMT if one were to use two start times?
I trade FMT at 6:30 only but use a TP=28 and SL=40 with no BE as per Wiseambitions and have found it to be profitable, but if adding a 6:15 start time to the trade would substantially increase the pips gained I would be willing to take the risk of a double entry and exposure with this system. Any comments would be welcome.

Bobby

Here are my back testing results using imported data from http://www.forextester.com/data/datasources.html
You can do your own back testing and check my results.



Your settings back tested at 6:30
6:30 UK time, SL=40, TP=28, No BE, Test period: 08-02-2010 to 12-31-2010: 68 trades, +409.7 pips,
6:30 UK time, SL=40, TP=28, No BE, Test period: 11-08-2010 to 12-31-2010: 25 trades, +131.0 pips


Your settings back tested at 6:15
6:15 UK time, SL=40, TP=28, No BE, Test period: 08-02-2010 to 12-31-2010: 65 trades, +603.0 pips,
6:15 UK time, SL=40, TP=28, No BE, Test period: 11-08-2010 to 12-31-2010: 26 trades, +227.0 pips


Mark's settings back tested at 6:30
6:30 UK time, SL=40, TP=35, BE=20, Test period: 08-02-2010 to 12-31-2010: 68 trades, +526.0 pips
6:30 UK time, SL=40, TP=35, BE=20, Test period: 11-08-2010 to 12-31-2010: 25 trades, -30.7 pips


My settings back tested at 6:15 (still testing on a demo)
6:15 UK time, SL=40, TP=35, BE=20, Test period: 08-02-2010 to 12-31-2010: 65 trades, +580.9 pips
6:15 UK time, SL=40, TP=35, BE=20, Test period: 11-08-2010 to 12-31-2010: 26 trades, +234.0 pips




Remember these are good estimates. I would half the profits and double the losses.



Here are the real trading results that are the benchmarks:


Mark's real trading results at 6:30
6:30 UK time, SL=40, TP=35, BE=20, Test period: 08-02-2010 to 12-31-2010: 83 trades, +267.0 pips
6:30 UK time, SL=40, TP=35, BE=20, Test period: 11-08-2010 to 12-31-2010: 29 trades, -83.0 pips


My real trading results at 6:30
6:30 UK time, SL=40, TP=35, BE=20, Test period: 11-08-2010 to 12-31-2010: 32 trades, -148.0 pips



Anyone else have any settings they would like to share using my benchmarks?

 

[wiseambitions]

Here are my back testing results using imported data from http://www.forextester.com/data/datasources.html

Remember these are good estimates. I would half the profits and double the losses.

So true. I have found plenty data sources which might indicate that

1) some members, especially Mark might have opened trades when it would not have happened for me

2) TP was touched but my broker did not close me out with profit

3) I would not consider SL had been touched according to the data, yet I was knocked out of the trade with a loss

The difference between TP and SL could be as much as 75 so not many such experiences are required in order to confound the data substantially. You might expect these marginal events to cancel themselves out but Murphy's law usually seems to prevail.

It doesn't just happen easily that you can make profit in FX! I treat a lot of backtesting with extreme caution.

 

[GumRai]

Hi, I'm litte bit confused about trading time you published. According to TMT manual, page 5, 6 all trade are checked(executed) on 5.30 GMT. Alpari UK ( im also using them ) operating GMT +1hour. And at the moment in UK ( UK time ) =GMT time. So wiseambitions , your trades ( If you operate from Alpari UK)should be executed on 5.30GMT( which at the moment is UK time) or 6.30 am Alpari UK. Do not be offended, I just want explanation cos im confused.
Cheers

I don't have TMT, so I can't comment on the start time with that.
FMT start time is 06:30 UK time

This thread is dedicated to FMT, there is another thread that discusses TMT

 

[PipZombie]

not sure what your point is, i was just joining to read the thread,whats wrong with that

Nothing wrong. Just watched the video and the link was nothing that I could see was about trading. If it was for entertainment purposes, I get it...other than that welcome to the forum.

 

[PipZombie]

How do you modify the numbers in the FMT EA? Every time I change something and close the window and open it back up the changes are not there.

I dont know if you have got your answer. But I posted a similar response for another user on post# 5222. Check it out and see if it answers your question. If not send me a Private Message and I can go into more details and even set up a skype session to go over everthing with you.

 

[billcurry]

So I retried getting the high-quality tick data from Dukascopy tonight and succeeded... and the results are amazing. I was concerned that higher quality historical data would kill my profits. After several hours playing with the settings for TMT, I couldn't have been more wrong...

Here is the backtest for 2010. Take note of the 99% modeling quality. You'll also be able to see my settings there.

From 2010.02.01 to 2010.12.01 with an initial deposit of $1,000:


Turned a $1,000 account into $734,096.00 in under a year with only 10% DD and 90% wins. Add to that the rebates I'd get from cashbackforex, and that would be nearly $800K for me. I know that there is a ridiculous amount of risk/trade, but honestly I think this EA is one of the few that it makes sense with.

These same settings for 2009 don't trade as well, but it still made over $300k!

Even you backtesting naysayers have to admit that these results are pretty interesting, and even half the profits would still be an amazing year.

P.S. I think we can all relax over the mediocre results this January -- I have been seeing similar results for the past few years. Things should start to turn around in February.

P.P.S. Follow this tutorial to get the same historical data. If you use this data keep in mind that the GMT offset is based on the data, not your broker. This data runs at GMT time, so there is no offset.

 

[RedGuerrilla]

RedGuerrilla,

Very interesting post and backtest results !! thanks for your input !

Actually you give me more credit than I deserve for my "research". My research was only running these start times for 5 weeks in October/ November on DEMO with 2 different brokers. The 6:15 and 6:30 start times gave very near the same results and were the best, but the other 2 times were still productive enough to supplement FMT net profits, so I kept them running for my LIVE accounts strategy that I run now. My research did NOT include back testing, I simply do not believe that backtesting is a good judge of how a EA will perform in live trading, there are too many variables, broker feed, spreads, slippage, etc., I know alot of you that love backtests will disagree, but this is my opinion for what it is worth ? So I find demo forward testing a better indication of an EA's potential and even then it is not perfect.
You are correct that the 6:15 and 6:30 start times are the best 2 but my live results show 6:30 slightly better than 6:15, then 6:00 is a little better than 6:45, I will drop any of the times if they fail to make a monthly profit but so far they are all doing well although January looks to be a challenge.

Anyway, RedGuerrilla, thanks for your very informative post !

cheers !

I agree with you I don't like back testing either. I will only do back testing if I can compare it to real trading results, thus creating a benchmark.

It's just a good estimate, but its an estimate with a reference. In other words, once I can verify that the data I'm using gives me a good enough estimate, than I can test other systems and see how good they are. At least I have a reference, it's better than nothing.

I'm trying to find systems that are worth doing forward testing. That way I'm not shooting in the dark and wasting my time. I was wasting my time on that 5 trade system. The qualified back testing showed some real bad results. I don't really have to have very accurate results to know that I'm going the wrong way.

I just have to look at the graph and get a good enough picture to determine if the system has better probabilities than another systems.

If the back tested results for each system are very close to each other, than it really doesn't matter because in real trading anything can happen.

But you must use good data. You can't just download the default historical MetaTrader data, or you'll get junk. You even get a warning about the data from the broker.

At lest I qualified my back testing by getting good data and comparing the results to real trading results so I can see close of an estimate it is to the real thing.

Here's where I learned to back testing correctly . . .

. . .the free downloadable data from MetaTrader (accessible via the Download button) is not always complete, and can contain large gaps. You can download free M1 data from http://www.forextester.com/data/datasources.html

From http://www.easyexpertforex.com/backtesting-optimization.html


I'm happy to help, I've really learn a lot from this forum with all the interaction. It's been really great. Thanks.

 

[RedGuerrilla]

'Live' Accounts:

01-20-2011, FMT 4.2, Short/Profit, FXSolutions, dealing desk, 1.5956, 1.5920, +36.0
My stats since 11-08-2010: 11 No Trades, 14 Break Evens, 16 Losses, 13 Profits, Total: -200.0 pips
Pips include slippage, using EA standard settings: 40 SL, 20 BE, 35 TP

01-20-2011, FMT 4.2, Short/Profit, IBFX, non-dealing desk, 1.59550, 1.59191, +35.9
My stats since 01-04-2010: 4 No Trades, 3 Break Evens, 3 Loss, 2 Profits, Total: -53.5 pips
Pips include slippage, using EA standard settings: 40 SL, 20 BE, 35 TP



Demo Accounts:

01-20-2011, FMT 4.2, 6: Short/Profit at non-dealing desk brokers:
PFG Best, 1.59552, 1.59202, +35.0
FOREX.com, 1.59551, 1.59199, +35.2
MB Trading, 1.59566, 1.59213, +35.3
FXCM, 1.59556, 1.59201, +35.5
Alpari(US), 1.59556, 1.59205, +35.1
FXDD, 1.59546, 1.59195, +35.1



Testing new tweaked settings, based on back testing. For more details, see Post #5659, page 708:
01-20-2011, FMT 4.2 EA, GU M15 Chart, UK time zone.
Using PFG Best Demo Account, Settings: SL=40, TP=30, BE=20, BE+5 pips,

6:15 start, sell trade, 1.59527, 1.59224, hit TP= +30.3 pips

for today +30.3 pips ...
My stats since 01-20-2011: Total: +30.3 pips



Hey billcurry, thanks for that back testing on the TMT Post #5710, impressive!

 

[trader578]

Nice posts Billcurry and RedGuerilla! You know how they say a picture is worth a thousand words - that's how I feel about backtests, especially those with 99% quality. Here's to a properous 2011 for all from this day forth!

 

[billcurry]

Okay guys, here is something pretty interesting. The last backtest I posted was with a $1,000 initial deposit. What if we were to split that $1,000 initial deposit into four $250 live accounts? Well, here are the results for one $250 account with the same settings as last time:


$250 to $605,183 in under a year, now that's pretty amazing.

But wait a minute, didn't we split that up into four accounts? But that would mean... around $2,420,732 total capital from the same $1,000. This is getting a bit ridiculous.

 

[Elovv]

What settings?
Best wishes
Elovv

Okay guys, here is something pretty interesting. The last backtest I posted was with a $1,000 initial deposit. What if we were to split that $1,000 initial deposit into four $250 live accounts? Well, here are the results for one $250 account with the same settings as last time:


$250 to $605,183 in under a year, now that's pretty amazing.

But wait a minute, didn't we split that up into four accounts? But that would mean... around $2,420,732 total capital from the same $1,000. This is getting a bit ridiculous.

 

[billcurry]

What settings?
Best wishes
Elovv

Check the backtest, they're all there. Using TMT.

 

[billcurry]

6:30 AM UK time, long signal, hit BE

 

[JWBear]

FXDD Long Signal at 1.59251 6:30 AM UK, no BE yet. Mine is set to 20.

 

[JWBear]

FXDD Long Signal at 1.59251 6:30 AM UK, no BE yet. Mine is set to 20.

6:15 AM was definitely they better start time tonight.

 

[PipZombie]

6:30 AM UK time, long signal, hit BE

what was your start-time and settings for TP SL BE ? thanks