Forex Skyline - boring but profitable!

[⭐ fxknight]

Hello,

I am sharing the Forex Skyline that I got from another forum which has a decent forward test. It apparently trades the USDJPY M15 and the forward test is here:

http://www.mt4i.com/users/atala/stats

 

Not an inflated return EA but nice and steady. I figured it would a welcomed change especially after so much Primeval and ZeusFX commotion lately (both useless than thrash IMHO).

 

Download: http://www.multiupload.com/E4IQV935BI

 

It is put out by the same developer of Reservoir Pips. Includes English manual and indicator. Please share your results/thoughts with us after you take it for a spin.

 

Cheers

[eric07]

4shared pls?

[hatchsead]

fxknight, i suppose it will be better with a broker with tight spread on that pair. If i may ask, what is the name of the broker you have it running on and the leverage?

[⭐ fxknight]

First this is not my forward test. It is the developer's I believe. Second, of course the tighter the spread the more money you can put in your pocket. However, this EA is not spread sensitive since it is looking for ~12 pips TP on every trade. For a good broker comparison I recommend to look at: http://www.fxintel.com/live/ and choose your broker there. I would stay away from IBFX, FXCM, Forex.com, FXDD etc. the ones with the highest fixed spreads. These are the true retail garbage of this industry and with the many ECN alternatives that are out there now there should be plenty of good choices. I gather that this live report is on Alpari UK. Hope this helps everybody ;)

 

Cheers

[⭐ fxknight]

4shared bans the links pretty quickly this is why I stay away from them. If you have issues downloading let us know and we can put it on slil.ru server.

[krisware]

Hello,

I am sharing the Forex Skyline that I got from another forum which has a decent forward test. It apparently trades the USDJPY M15 and the forward test is here:

http://www.mt4i.com/users/atala/stats

 

Not an inflated return EA but nice and steady. I figured it would a welcomed change especially after so much Primeval and ZeusFX commotion lately (both useless than thrash IMHO).

 

Download: http://www.multiupload.com/E4IQV935BI

 

It is put out by the same developer of Reservoir Pips. Includes English manual and indicator. Please share your results/thoughts with us after you take it for a spin.

 

Cheers

 

Maybe I could add MM to it and make it more profitable.

[syk1kz]

Thanks for the time you took to post this system and the stats along with your comments.

 

 

I have NOT downloaded the EA but just from looking at the stats on the mt4i link i can see big issues with those settings being used.

No matter what you do with a 50SL and 18TP you are destined to fail sooner or later, its like megadriod all over again.

 

Lets just take a random looser from the mt4i link above shall we - 42839313 2010/05/18 16:53:00 Buy 0.20 USDJPY 92.681 92.12 92.87 2010/05/18 22:05:00 92.12 0.00 0.00 - 121.80 - 56.1

 

so -56.1pips is the average stoploss, some are 54 some are 58 lets just go with 54. i can see from looking on the stats, thats a risk:reward or 3:1

 

So if we win 3 trades in a row [ TP is 18-20pips looking at those stats ] then we have bagged 57pips [ we will go with 19TP ] wow nicely done for an EA i will admit.

Then we follow up with a single loosing trade of 54pips and we are back to square one. [ ok we have 3 pips net, time to party ]

Now we take another Loss for whatever reason - we suddenly need 3 winners just to breakeven, this hardly seems prudent and beneficial to me.

 

If the above poster was willing to add MM then i would retract 90% of this comment, but then its not the same EA anymore is it its been modified somewhat.

 

anyway the point is becarefull with bad money managenment robots, they might look ok this month, what about next month and the month after ?

 

and i just noted its only made 81pips in 3 months? those stoplosses sure are hurting those stats :|

[⭐ fxknight]

This is a backtest on USDJPY M15 from 01-2009 til current dafault settings. Looks very decent with small draw. Seems to have a "recovery" feature built into its MM which can be improved of course.

 

http://www.forumimagecodes.com/images/0yozq2aozl94erzmqvh.jpg

 

http://www.forumimagecodes.com/images/dlsg45k3ftm2jq88xih.jpg

Edited August 17, 2010 by fxknight

[⭐ fxknight]

Thanks for the time you took to post this system and the stats along with your comments.

 

 

I have NOT downloaded the EA but just from looking at the stats on the mt4i link i can see big issues with those settings being used.

No matter what you do with a 50SL and 18TP you are destined to fail sooner or later, its like megadriod all over again.

 

Lets just take a random looser from the mt4i link above shall we - 42839313 2010/05/18 16:53:00 Buy 0.20 USDJPY 92.681 92.12 92.87 2010/05/18 22:05:00 92.12 0.00 0.00 - 121.80 - 56.1

 

so -56.1pips is the average stoploss, some are 54 some are 58 lets just go with 54. i can see from looking on the stats, thats a risk:reward or 3:1

 

So if we win 3 trades in a row [ TP is 18-20pips looking at those stats ] then we have bagged 57pips [ we will go with 19TP ] wow nicely done for an EA i will admit.

Then we follow up with a single loosing trade of 56pips and we are back to square one.

Now we take another Loss for whatever reason - we suddenly need 3 winners just to breakeven, this hardly seems prudent and beneficial to me.

 

If the above poster was willing to add MM then i would retract 90% of this comment, but then its not the same EA anymore is it its been modified somewhat.

 

anyway the point is becarefull with bad money managenment robots, they might look ok this month, what about next month and the month after ?

 

and i just noted its only made 81pips in 3 months? those stoplosses sure are hurting those stats

 

 

I fully agree, however it could hurt if you have many losers in a row. This EA has pretty good stats I think as it is and you can always reduce the SL to 2:1 or even 1:1 to see what happens. Not trying to defend this EA but compared to what we see here day in and day out it is a step in the right direction nevertheless. Everybody needs to play with it and tailor it for their custom trading style if they plan to use it live in the future. Its a good start anyway.

 

Cheers

[halcyonn]

FxKnight sir.i got this in the journal during backtes which is hallmark of martingale....

 

 

2010.08.17 16:24:54 2009.03.09 07:00 Tester: not enough money for buy 32.00 USDJPY at 98.036 sl: 0.000 tp: 0.000 [2009.03.09 07:00]

 

also in july 2008 with 25 pip tp and 55 pip sl settings

 

2010.08.17 16:39:35 2008.07.14 16:45 Tester: not enough money for buy 32.00 USDJPY at 106.490 sl: 0.000 tp: 0.000 [2008.07.14 16:45]

Edited August 17, 2010 by halcyonn

[⭐ fxknight]

Its not martigdale.. it is its recovery method. If it loses it increases the lots on the next few trades but does not load up like martingdale. If you run it with large lot size on a small account this will happen. Run it with the default 0.02 lots on a $5k account and you won't have this issue. You can also try TP of 80 instead of 55 as default. Seems to do well with higher TP as the stats become more balanced.

[krisware]

Thanks for the time you took to post this system and the stats along with your comments.

 

 

I have NOT downloaded the EA but just from looking at the stats on the mt4i link i can see big issues with those settings being used.

No matter what you do with a 50SL and 18TP you are destined to fail sooner or later, its like megadriod all over again.

 

Lets just take a random looser from the mt4i link above shall we - 42839313 2010/05/18 16:53:00 Buy 0.20 USDJPY 92.681 92.12 92.87 2010/05/18 22:05:00 92.12 0.00 0.00 - 121.80 - 56.1

 

so -56.1pips is the average stoploss, some are 54 some are 58 lets just go with 54. i can see from looking on the stats, thats a risk:reward or 3:1

 

So if we win 3 trades in a row [ TP is 18-20pips looking at those stats ] then we have bagged 57pips [ we will go with 19TP ] wow nicely done for an EA i will admit.

Then we follow up with a single loosing trade of 56pips and we are back to square one.

Now we take another Loss for whatever reason - we suddenly need 3 winners just to breakeven, this hardly seems prudent and beneficial to me.

 

If the above poster was willing to add MM then i would retract 90% of this comment, but then its not the same EA anymore is it its been modified somewhat.

 

anyway the point is becarefull with bad money managenment robots, they might look ok this month, what about next month and the month after ?

 

and i just noted its only made 81pips in 3 months? those stoplosses sure are hurting those stats

 

 

I fully agree, however it could hurt if you have many losers in a row. This EA has pretty good stats I think as it is and you can always reduce the SL to 2:1 or even 1:1 to see what happens. Not trying to defend this EA but compared to what we see here day in and day out it is a step in the right direction nevertheless. Everybody needs to play with it and tailor it for their custom trading style if they plan to use it live in the future. Its a good start anyway.

 

Cheers

 

The Idea of this EA is TP = SL and every loosing lot will be doubled - so let say You had 4 looser in a row - this would mean You would loose (2^3+2^2+2^1+2^0)*FixedLot * PipsSL/TP = (8 + 4 + 2 + 1) * FixedLot * PipsSL/TP

So next lot size would be (2^4+2^3+2^2+2^1+2^0)*FixedLot * PipsSL/TP = (16 + 8 + 4 + 2 + 1) * FixedLot * PipsSL/TP

 

Because 2^4 = 16 is larger than (2^3+2^2+2^1+2^0) = 15 - so If winning bid will come up it would restore all losses and go further.

 

This idea is very simple and it uses probability - how big is chance loosing let say 5 or 6 or 10 times in a row.

 

Same Idea like ProForexRobot with 4 other pairs.

[krisware]

I'm going to build one EA which uses same Idea and no indicators - only probability and Random numbers (with indicators later).

Anyway I have used this so called recovery system before with Forex Shocker 1.1 and few others and got good results with backtesting, but I'm not using it live, yet (bit risky), but I'm going to develope NewsFilter and few more indicators and it should be good enough.

[Stormin_Norman]

fxknight. do you not remember discussing this with me months ago :)

 

it was a free EA from some other garbage commercial system. this 'free' EA turned out better than the commercial one! however using its martingale positioning leads it to blow accounts over 10 year tests.

 

running it at 20% risk over 10 years returns about 70% during those 10 years.

 

did you bump your head causing you to start looking at martingales? ;)

Edited August 17, 2010 by Stormin_Norman

[⭐ fxknight]

Its not martingdale per se in the way other martingdale EAs are. EA Martingdale definition is when an EA accumulates positions against an infavorable market. This EA takes larger positions when it has lost - after the losses are closed in an attempt to recover those losses. Big difference.. actually night and day. Strictly from a gambling point of view when it "doubles up" after a loss could be called martingdale. I ran a BT on this EA from 1999 - current and it survived nicely. Give it a shot with the default settings on a $10k account and you will see. The DD I got was 17% for a 60% plus gain. Edited August 17, 2010 by fxknight

[noone]

So are you using it live fxknight?

[halcyonn]

Its not martingdale per se

 

 

quite true but its recovery sizing is awry and that gives an impression of martingale.

[Stormin_Norman]

Its not martingdale per se in the way other martingdale EAs are. EA Martingdale definition is when an EA accumulates positions against an infavorable market. This EA takes larger positions when it has lost - after the losses are closed in an attempt to recover those losses. Big difference.. actually night and day. Strictly from a gambling point of view when it "doubles up" after a loss could be called martingdale. I ran a BT on this EA from 1999 - current and it survived nicely. Give it a shot with the default settings on a $10k account and you will see. The DD I got was 17% for a 60% plus gain.

 

it is martingale. like martingale betting it loses a 'bet' so it opens the next one at double the size. the only difference is it closes the previous trade before entering a new one.

 

roughly a 20% DD for a 5-6% per year gain isn't all that crash hot. banks here offer 6%+ for term deposits. but it does show that not all martingales do crash accounts; even if to do so their returns are very small.

[krisware]

it is martingale. like martingale betting it loses a 'bet' so it opens the next one at double the size. the only difference is it closes the previous trade before entering a new one.

 

roughly a 20% DD for a 5-6% per year gain isn't all that crash hot. banks here offer 6%+ for term deposits. but it does show that not all martingales do crash accounts; even if to do so their returns are very small.

 

I'v tested it with 0.01 lot size with 500 USD deposit and it makes about 500-1000 USD/y. Max DD is 31*LotSize * TP. It could be higher, but backtests of 10 years show that it is ok. With 55 pips TP/SL max loosing amount would be ~170 USD. It is much safer than avg. Martingale and You may use it on lower deposits, too.

[Stormin_Norman]

over 10 years on 0.01 maximum drawdown was 2,500. 5 times your starting balance. sometimes it is tricky to fully read a backtest statement.

 

hence i said 25% drawdown with a 70% profit. over the corse of 10 years at 0.01 it makes about $7,000.

[vikram88]

Even Megadroid has a recovery mode

However I think adding a Trailing Stop to this EA would improve its performance

Can someone please modify and share

Thanks

[krisware]

I'm going to build one EA which uses same Idea and no indicators - only probability and Random numbers (with indicators later).

Anyway I have used this so called recovery system before with Forex Shocker 1.1 and few others and got good results with backtesting, but I'm not using it live, yet (bit risky), but I'm going to develope NewsFilter and few more indicators and it should be good enough.

 

Now it is ready and I'm backtesting it with 4 Digit broker, with starting amount 5000K with 0.01 as min lot and 55 TP/SL.

AND as I told You before it works same way as I thought - but because it uses just Rundom numbers it was loosing max 6 times in a row at best.

I had to find better RandomSeed number for Backtesting, because LocalTime() is not enogh - so I used MathSrand(TimeLocal()+MathRound(Bid * Ask * 1000) + TimeCurrent());

bool dir = MathRand() >= 16384;

 

Good thing is - that it trades all the time - If trade is closed -> new random directed trade is opened.

 

I'll post this simle EA as soon as I'v tested it few times.

 

Anyway this EA I made just for proving that probability works in best case, but it could blow Your account - More like playing in Casino.

 

1. test blowed account because bad RandomSeed (or bad luck) - I'v checked that had 2 lot limit on micro so account was not blown - it was not able to complete 2.56 lot order (my mistake)

2. test starting from 2009.01.02 till 2009.05.06 -> account was growing from 5k to 6544. Profit 1544 with Max DD 340 USD. -> finally it had same problem like fist one - 2 lot limit reached.

I'm switching to Alpari with 500 TP/SL (5 digit) 2009.01 - 2010.05.

1. 50k account + 3,3 k profit DD 16.8 % (8865 USD) max 10 looser in a row

 

 

Lowering TP showed that probability of loosing more than 10 times in a row was growing and finally it proved to be more like in Casino - using just luck is not enough. House always wins.

Higher TP works better, but i think some day it could have that 10 in row loosers, too. And as told before Higher TP means less profit - so You could earn more just sitting and earing interest with 50k in any bank.

 

I'm going to use few indicators and filters next time, but I think Forex is Casino in a long run with this kind of Strategy - You need to learn how to use News, patterns, etc.

 

GBPUSD had better results than EURUSD with Alpari, 2009-2010.05, 10k dep +3,8k profit, DD was 15.22% (1624 USD), TP 800 (5 digit br)

Edited August 18, 2010 by krisware
conclusion

[⭐ fxknight]

So are you using it live fxknight?

 

No I am not using it live. Too risky for my taste. If I can decrease the risk I might give it a shot but for now I stick to "one position at a time" EA's with no added risk following losses.

 

Cheers

[hyperdimension]

fxknight. do you not remember discussing this with me months ago :)

 

it was a free EA from some other garbage commercial system. this 'free' EA turned out better than the commercial one! however using its martingale positioning leads it to blow accounts over 10 year tests.

 

running it at 20% risk over 10 years returns about 70% during those 10 years.

 

did you bump your head causing you to start looking at martingales? ;)

Skyline was discussed in the PipsReservoir thread almost 4 months ago: http://indo-investasi.com/showthread.php/6120-Forex-PipsReservoir?p=72087&highlight=skyline#post72087

 

It actually looked like a good martingale system. It seems like everyone just ignored my posts and the EA at the time.

[⭐ fxknight]

Skyline was discussed in the PipsReservoir thread almost 4 months ago: http://indo-investasi.com/showthread.php/6120-Forex-PipsReservoir?p=72087&highlight=skyline#post72087

 

It actually looked like a good martingale system. It seems like everyone just ignored my posts and the EA at the time.

 

The fact that it survives the backtests from 1999 it put it in a category above any other EA out there that is based on increasing lotsize (and risk). With few mods I think it can be used for live trading.

 

Cheers