Scalping the EUR/GBP effectively - Forturama | Page 2 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

How to use the 123 strategy effectively 1,683 replies

How to use the 9/20 EMA setup effectively 162 replies

How To Effectively Trade In Both Range And Trending Market By A Single Method 39 replies

How does one scalp the eurusd effectively? 13 replies

Attachments: Scalping the EUR/GBP effectively - Forturama
Exit Attachments

Scalping the EUR/GBP effectively - Forturama

  • #21
  • Sep 17, 2008 11:25am Sep 17, 2008 11:25am
so you found the holy grail and share it here!

Thanks a lot!

Quoting Forturama
Ok,

Now we have to be sure that this history about RSI is really true.

I create an EA to simply buy and sell on RSI OS and OS zones.

I took 5 pips as TP and 10 pips as SL.

NO MARTINGALE IS USED. JUST TESTING RSI USABILITY.

The results are amazing.

See attached files.

Fixed Lot size = 1

INITIAL Deposit : USD1000

PERIOD: From 2004 to Last month
  • #22
  • Sep 17, 2008 11:33am Sep 17, 2008 11:33am
Honestly, ur system rocks by RSI's reliability during a quiet Asian hour, NOT by a sophisticated MM algorithm.

Forget about Martingale, it won't help ur system a lot but may end up blow up it.

btw, Martingale is about keep doubling (simply put) multiple times, but ur trade only last during Asian hours, you can at most double it for 2 consecutive times? then Asian session is over, what do u do then?

Quoting Forturama
Hi all,

I found some where a trade system that uses semi-martingale.
  • #23
  • Sep 17, 2008 11:41am Sep 17, 2008 11:41am
I used to be like u.

but been there a few times (margin call), I find the answer to my question.

the question is: what's so difficult to use a stoploss?

the answer is: nothing in fact.

so I'd better go with a SL when things go bad.

Answer me this:

Does your EA apply no Stoploss too? and reach over 60k from 1k @ 2004 till now?

Thanks.

Quoting Forturama
It is up to you.

Personally I have no stoploss, my stop is margin call.
  • #24
  • Sep 17, 2008 11:43am Sep 17, 2008 11:43am
Good question...

I do not know what to do.

Pehaps this should be a sl rule.

I stop trading 6pm, if my positions are open 8pm I close it, positive or not.

Quoting charvo
Honestly, ur system rocks by RSI's reliability during a quiet Asian hour, NOT by a sophisticated MM algorithm.

Forget about Martingale, it won't help ur system a lot but may end up blow up it.

btw, Martingale is about keep doubling (simply put) multiple times, but ur trade only last during Asian hours, you can at most double it for 2 consecutive times? then Asian session is over, what do u do then?
{Promotion Removed}
  • #25
  • Sep 17, 2008 12:00pm Sep 17, 2008 12:00pm
You are right, I use stoploss on my ea, to test RSI.

The question is that 2008 is not a good year (see picture). So i decided to try some method to avoid losses.

I an not allergic to stoploss, I look for a good solution.

I will try use semi-martingale and set stoploss to 3%.

If the statistic is right even with an risk reward ratio like this I should have some profit.

What is your opinion?

Quoting charvo
I used to be like u.

but been there a few times (margin call), I find the answer to my question.

the question is: what's so difficult to use a stoploss?

the answer is: nothing in fact.

so I'd better go with a SL when things go bad.

Answer me this:

Does your EA apply no Stoploss too? and reach over 60k from 1k @ 2004 till now?

Thanks.
{Promotion Removed}
  • #26
  • Sep 17, 2008 12:05pm Sep 17, 2008 12:05pm
I have attached the settings file. TP=5, SL=19, trading hours 17 to 7 Alpari server time.
Quoting Forturama
could you tell us what parameters you are using?

time to trade, sl and tp?
Attached File(s)
File Type: zip eurgbp king.zip   < 1 KB | 1,016 downloads
  • #27
  • Sep 17, 2008 12:17pm Sep 17, 2008 12:17pm
Quoting Forexharry
Why don't you just learn how to trade period ? If you are doing this with real money it is only a matter of time till you bust your account. If you will gamble like this in Vegas they will send a plane to pick you up if you will bring a $100,000 there. Good luck...
and how about if YOU teach US to trade, huh?
  • #28
  • Sep 17, 2008 1:47pm Sep 17, 2008 1:47pm
Hi, Fortu,

if statistics are right, ur system is already a holy grail to me :-).

if it is really that simple only using RSI, then now all you need is to forward test it, either demo or live.

i do think of some filters that seem to "improve it", but in fact, these filters usually ruin things.

these filters include:

1. narrow ur trading session by cutting the beginning hour and noon hour of Asian time;

2. calculate historical average movement in Asian morning session (use that for SL) ;

3. one more indicator/calculation to monitor the size of price movement;
(or a range indicator)

Again, I think ur system is already great enough and need none of these. 2008 may not be a perfect year for it. that's it.

I will do some backtest of ur system later. will update.

Thank you again.


Quoting Forturama
If the statistic is right even with an risk reward ratio like this I should have some profit.

What is your opinion?
  • #29
  • Sep 17, 2008 2:25pm Sep 17, 2008 2:25pm
Thank you for your encouragement,

I tested the EA with 2008 data and the results are not so bad.

The DD is too high, starting from 10000 I went to 3000 and after I reached 30.000!

This means that I just have to believe the system and KISS: Keep It Simple, Stupid.

If you are interested I can send you a copy of my EA, it is very simple but it will save some minutes of coding.

I look for a solution to test an EA with pending orders (the second version with martingale).

See file with back test 2008.
Quoting charvo
Hi, Fortu,

if statistics are right, ur system is already a holy grail to me :-).

if it is really that simple only using RSI, then now all you need is to forward test it, either demo or live.

i do think of some filters that seem to "improve it", but in fact, these filters usually ruin things.

these filters include:

1. narrow ur trading session by cutting the beginning hour and noon hour of Asian time;

2. calculate historical average movement in Asian morning session (use that for SL) ;

3. one more indicator/calculation to monitor the size of price movement;
(or a range indicator)

Again, I think ur system is already great enough and need none of these. 2008 may not be a perfect year for it. that's it.

I will do some backtest of ur system later. will update.

Thank you again.
Attached Image (click to enlarge)
Click to Enlarge Name: StrategyTester_2008.gif Size: 8 KB
Attached File(s)
File Type: zip StrategyTester_2008.zip   29 KB | 404 downloads
{Promotion Removed}
  • #30
  • Sep 17, 2008 3:36pm Sep 17, 2008 3:36pm
For this kind of performance, one MUST do trial and error before going LIVE.

it might be one reason:

for example:

3 yrs ago, eurusd is only 1.2000

so average daily movement maybe was around 35 pips.

now, eurusd is 1.5000, so average daily movement maybe was around 45 pips. (consider gbpusd, 1.8000, daily move average maybe 85 pips.....)
_____________________________________________________________
my point is that maybe 6 or 7 pips instead of 5, will be significant to the results. Mathmatically, the change from 5 to 6 is 20% increase. however, when incorporate probabilities, we need to further look into it.

5 pips TP seems great for 2004~2007, but something of the market remain same while the others change. we can find it out.

Quoting Forturama
Thank you for your encouragement,

I tested the EA with 2008 data and the results are not so bad.

The DD is too high, starting from 10000 I went to 3000 and after I reached 30.000!

This means that I just have to believe the system and KISS: Keep It Simple, Stupid.
  • #31
  • Sep 17, 2008 4:25pm Sep 17, 2008 4:25pm
Quoting Forturama
Thank you for your encouragement,

I tested the EA with 2008 data and the results are not so bad.

The DD is too high, starting from 10000 I went to 3000 and after I reached 30.000!

This means that I just have to believe the system and KISS: Keep It Simple, Stupid.

If you are interested I can send you a copy of my EA, it is very simple but it will save some minutes of coding.

I look for a solution to test an EA with pending orders (the second version with martingale).

See file with back test 2008.
This looks intresting. I would like to have that EA, if you wanna send it to me.
  • #32
  • Sep 17, 2008 7:23pm Sep 17, 2008 7:23pm
This EA causes leathal error to my system. cannot run it.

Quoting Forturama
Normally I am sleeping during Asian (I am Brazilian but I live in France).

Before create an EA I tested this strategy manually and I used my RSI alert indicator. It sounds every time RSI reaches a selected value.

See file.
  • #33
  • Sep 18, 2008 12:47am Sep 18, 2008 12:47am
What EA?

I did not send it yet.

I sent an expert RSI_ALERT.

Quoting charvo
This EA causes leathal error to my system. cannot run it.
{Promotion Removed}
  • #34
  • Sep 18, 2008 12:49am Sep 18, 2008 12:49am
Are you talking about parameters optimization?

Quoting charvo
For this kind of performance, one MUST do trial and error before going LIVE.

it might be one reason:

for example:

3 yrs ago, eurusd is only 1.2000

so average daily movement maybe was around 35 pips.

now, eurusd is 1.5000, so average daily movement maybe was around 45 pips. (consider gbpusd, 1.8000, daily move average maybe 85 pips.....)
_____________________________________________________________
my point is that maybe 6 or 7 pips instead of 5, will be significant to the results. Mathmatically, the change from 5 to 6 is 20% increase. however, when incorporate probabilities, we need to further look into it.

5 pips TP seems great for 2004~2007, but something of the market remain same while the others change. we can find it out.
{Promotion Removed}
  • #35
  • Sep 18, 2008 2:54am Sep 18, 2008 2:54am
I send you the .ex4 file of an EA called RSI_PRINCE inspired on EURGBP_KING, created by my Brazilian colleague and discussed on an Internet forum (tsd i guess).

The parameters are self explained.

Please, let me know if you have good results wiht other parameters.

I run an optimization with start time, stop time, stoploss and take profit.

ATTENTION:
The variable RISK is not used.
When MM = true the lot size is free margin/1000 with minimal value = Lots.

The AE opens an order when RSI reaches OS ou OB zone. I would like it to open an order when it is exiting those zones.


Any body has an idea of how to do this?
Attached File(s)
File Type: ex4 RSI_PRINCE.ex4   4 KB | 1,022 downloads
{Promotion Removed}
  • #36
  • Edited 10:06am Sep 18, 2008 4:21am | Edited 10:06am
The 2008 optimization is finished.

The results very good.

There is a "philosophical" change: To improve profits we have to increase SL and TP. So th EA is no more scalper.

Without Optimization:

TP = 5
SL = 10

With Optimization:
TP = 21
SL = 16

To maintain the same risk level I calculated the Lot size that represents the same % of loss with different Lot sizes. So:

SL = 10 ---- Lot = 1
SL = 16 ---- Lot = 0.63

See attached comparison between optimized and not optimized backtest:
Attached Image(s) (click to enlarge)
Click to Enlarge Name: StrategyTester_2008_not_optimized.gif Size: 8 KB
Click to Enlarge Name: StrategyTester_2008_optimized.gif Size: 7 KB
{Promotion Removed}
  • #37
  • Sep 18, 2008 4:47am Sep 18, 2008 4:47am
Now the first 16 days of September (from 1 to 16)
Attached Image (click to enlarge)
Click to Enlarge Name: 16_9.gif Size: 8 KB
{Promotion Removed}
  • #38
  • Edited 6:57am Sep 18, 2008 6:44am | Edited 6:57am
An important detail: All ALPARI data is CET and not GMT.

I am back testing the EA with ALPARI DATA.

http://en.wikipedia.org/wiki/Central_European_Time
{Promotion Removed}
  • #39
  • Sep 18, 2008 11:51am Sep 18, 2008 11:51am
This is very unexpected. Surprising.

I was not talking about optimization actually, i guess that maybe sensitivity test of 6,7,8 instead of 5 pips take-profit is what i mean. only do sensitivity test for the TP and SL, or RSI's 30, 70.

but this fundamentallly change to 21, 16 in fact convert this EA to another philosophy as you said. we'd better be conservative on that, since optimization seems not great for future (just my impression)

i'll test it after getting home.

Quoting Forturama
The 2008 optimization is finished.

The results very good.

There is a "philosophical" change: To improve profits we have to increase SL and TP. So th EA is no more scalper.
  • #40
  • Sep 19, 2008 3:01am Sep 19, 2008 3:01am
A good trading night.

Two opportunities.

We let money on the table.

Take profit of 5 pips is perhaps too short.

See pics.
Attached Image (click to enlarge)
Click to Enlarge Name: eurgbp_19092008-01.gif Size: 18 KB
{Promotion Removed}
0 traders viewing now