50 pips candle system | Page 29 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

MT4 candle-by-candle manual backtesting 9 replies

How to make EA that send Open Price of Candle for every new candle 5 replies

Day's first H4 candle correlation to daily candle 14 replies

Correlation between first 4 hour candle and daily candle 1 reply

Is there a programmer interested to make an ea for 50 pips candle system? 0 replies

Attachments: 50 pips candle system
Exit Attachments

50 pips candle system

  • #561
  • Dec 16, 2009 12:19pm Dec 16, 2009 12:19pm
Quoting zebulon
i'm agree with that.

Is it possible for you C-Speed to modify your tester and allow us to test also with 100 pips ?? It would be very great...
I will add 25 and 100 pips to the tester, however due to the complexity and lack of time, I cannot make the tester work for custom pip ranges.

Give me a few minutes.
  • #562
  • Dec 16, 2009 12:41pm Dec 16, 2009 12:41pm
Ok, 25, 50, 100, and 200pip option on GBP/USD is now available to back test.
  • #563
  • Dec 16, 2009 12:56pm Dec 16, 2009 12:56pm
Quoting sangmane
Hello Kostas,

Pls find 50 pips EA below, it seems using MA as filter give improvement in result. Adding Zebulon's evolution, almost double the account in 1 year. Unfortunately, it's only in backtesting lol.
Your CCI filter seems contra productive. Pls experiment, tweak the setting, and post your result.

Hands in your balls, ooppss, ball in your hand now...
Great job man. Tell me please if I want the ea martingale to go in this sequense 0.1, 0.2, 0.4 witch style I have to choose?Thank you
  • #564
  • Dec 16, 2009 1:13pm Dec 16, 2009 1:13pm
hi there,
can you post up the setting you make for it? thanks

Quoting sangmane
Hello Kostas,

Pls find 50 pips EA below, it seems using MA as filter give improvement in result. Adding Zebulon's evolution, almost double the account in 1 year. Unfortunately, it's only in backtesting lol.
Your CCI filter seems contra productive. Pls experiment, tweak the setting, and post your result.

Hands in your balls, ooppss, ball in your hand now...
  • #565
  • Edited 1:44pm Dec 16, 2009 1:14pm | Edited 1:44pm
Did some quick analysis with my back tester.

It looks like using 25pips produces the best results!

Using these sequences:
GGR
RRG
GRGR
RGRG

This month +225pips
The last 6 months +7,175pips!
From 1/1/2009 +4,300pips

Using original:
GGG
RRR
GRGR
RGRG

This month -200pips
Last 6 months +3,000pips
1/1/2009 + 3,300pips


*EDIT: Miscalculation in the number of pips has been fixed. I've updated this post with the correct results.
  • #566
  • Dec 16, 2009 1:25pm Dec 16, 2009 1:25pm
Hi C-speed,
Do you play risk or martingale? Or just fix lot for each trade?


Quoting C-speed
Did some quick analysis with my back tester.

It looks like using 25pips produces the best results!

Using these sequences:
GGR
RRG
GRGR
RGRG

This month +450pips
The last 6 months +14,350pips!
From 1/1/2009 +8,600pips

Using original:
GGG
RRR
GRGR
RGRG

This month -200pips
Last 6 months +3,000pips
1/1/2009 + 3,300pips
  • #567
  • Dec 16, 2009 1:26pm Dec 16, 2009 1:26pm
Quoting special0nefx
Hi C-speed,
Do you play risk or martingale? Or just fix lot for each trade?
Fixed lot.
  • #568
  • Dec 16, 2009 1:27pm Dec 16, 2009 1:27pm
The reason for that is the lenght of time/number of trades. A better comparison would be to compare by the same number of transactions rather than a date range. If theory has it right, this system should shield the same results, regardless of the pip range. If you choose to use 100, 200 or 1000 pips, it's the same. The higher the pip range, the longer you would have to wait to see results. I bet that if you backtest by 25 pips you would have loosing weeks just as the current system has loosing months. I guess it depends what type of trader you are or want to be. A martingale seems to be the best fit for this system and although most people are terrified of martingale, this type of system are the ones that better apply to the concept since your losses are distributed among your wins and the 50/50 nature of the candles/trade direction change makes it very difficult to have a train of loses that is too long to sustain. It is not impossible, but I think with good money management and a way (indicator or similar) to tell us where the market might be going, then it might not be such a bad idea.
Please pull my head out of my ass, I am running out of oxygen...
  • #569
  • Dec 16, 2009 1:30pm Dec 16, 2009 1:30pm
ok thanks. Nice sequences you found here...Will try it new modify sequences you found. thanks
Quoting C-speed
Fixed lot.
  • #570
  • Dec 16, 2009 1:35pm Dec 16, 2009 1:35pm
Quoting C-speed
Did some quick analysis with my back tester.

It looks like using 25pips produces the best results!

Using these sequences:
GGR
RRG
GRGR
RGRG

This month +450pips
The last 6 months +14,350pips!
From 1/1/2009 +8,600pips

Using original:
GGG
RRR
GRGR
RGRG

This month -200pips
Last 6 months +3,000pips
1/1/2009 + 3,300pips
the tester counts allways with 50 pips no matter what size you use, so the results will be smaller if you use 25
  • #571
  • Dec 16, 2009 1:40pm Dec 16, 2009 1:40pm
Quoting Saintsurfer
the tester counts allways with 50 pips no matter what size you use, so the results will be smaller if you use 25
Haha, oops! You're right.. I fixed it.

I edited the original post with the correct results.
  • #572
  • Dec 16, 2009 2:10pm Dec 16, 2009 2:10pm
BUT it is imposible to get simillar results on the ea and tester for now...
  • #573
  • Dec 16, 2009 2:34pm Dec 16, 2009 2:34pm
Quoting C-speed
Did some quick analysis with my back tester.

It looks like using 25pips produces the best results!

Using these sequences:
GGR
RRG
GRGR
RGRG

This month +225pips
The last 6 months +7,175pips!
From 1/1/2009 +4,300pips

Using original:
GGG
RRR
GRGR
RGRG
C-speed,
Can you explain the sequences? For example, when you mention GGR do you mean that after GG you sell trying for an R?
And how about GRGR, do you mean after GRG you sell for an R? (I am trying to match the EA's results with your numbers which is why I am asking).
  • #574
  • Dec 16, 2009 2:40pm Dec 16, 2009 2:40pm
Quoting kostas1
C-speed,
Can you explain the sequences? For example, when you mention GGR do you mean that after GG you sell trying for an R?
And how about GRGR, do you mean after GRG you sell for an R? (I am trying to match the EA's results with your numbers which is why I am asking).
You are correct. If you wish to enter LONG (BUY) when you see "GG", then the winning sequence would be "GGG" to use on the tester.

If you wish to enter SHORT (SELL) when you see "GRG", then you will enter "GRGR" in the tester.
  • #575
  • Dec 16, 2009 5:46pm Dec 16, 2009 5:46pm
Thank you C-speed !
  • #576
  • Dec 16, 2009 6:40pm Dec 16, 2009 6:40pm
Hello everybody,

I am new here, starting membership today.
I am from Brazil , so any english mistake should be forgiven..

Regarding your 50 pips method, I am starting to look at it and using the 15 min chart it seems to work well too. and it seems very profitable

I've been using something near that but with a little bit different method:
I use Parabolic SAR as the first referencial.

As soon as the first SAR point pops up, I subtract of it 50 pips or add 50 pips depending on what part of the chart it appears, and those are my Buy or sell limit order with 50pips TP, but as I have a little bit strong stomach I use 300 pips of SL and I can tell you that I am very...but very very very rarely stopped, however I work with mini-lots only in regular basis. To increase them I combine the SAR with RSI(14), but it only can be done using programming script with "if-else" routine.

When 60 <=RSI(14) <= 30 I increase the lots in the order.

As I use Meta4 (MetaTrader Plataform) I wrote a script that do that automaticlly...but it is still in test so as soon as I am sure it works properly I will post it here to share with those who uses the same plataform or other that allows to import scripts.

Ok then...see you around
  • #577
  • Dec 16, 2009 7:26pm Dec 16, 2009 7:26pm
Quoting C-speed
You are correct. If you wish to enter LONG (BUY) when you see "GG", then the winning sequence would be "GGG" to use on the tester.

If you wish to enter SHORT (SELL) when you see "GRG", then you will enter "GRGR" in the tester.
Thanks C-speed.

I am glad you changed your mind and made an option such that we could experiment with different pip ranges.


Now the question is can some one come up with a rule for placing or choosing the orders such that wins and losses can alternate, but a win can not be two losses away in 12 or 14 trades. For example

L L L L L L L L L W L W
1 2 3 4 5 6 7 8 9 10 11 12 is a good win. As another example

L L W L L L W L L W L W
1 2 3 4 5 6 7 8 9 10 11 12 is also acceptable since a Win is not more than two wins away from a loss within 12 trades.

If we have two Wins in a row (for example W W) at any point within the twelve trades we stop and go back and start trading again from One. Also at any moment if we have for example (W L W) we also stop and go back and start to trade from one again.

This system was very successfull in roullette and Backarat so much that the casinos knew us and wiil stop us from playing many times.

I have the Semi-martingale series written out for this. If we can successfully come out with a good trading method then I can release the series. However it is good to know that the Casino rules are some what different from FX.
  • #578
  • Dec 17, 2009 12:52am Dec 17, 2009 12:52am
Quoting C-speed
Thanks sangmane!

It looks like there is a problem with your EA. If you turn off Martingale system, the EA generates an error. I changed the lot size from 0.01 to 1 in order to fix the error.

However there still seems to be a discrepancy. If lot size is set to 1, you get 2 different results with Martingale on versus off. You should get the same results.

I'm a coder, but I'm not familiar enough with MT4 EA code to make an immediate fix.
Hi c-speed,

thanks for the info about the error. I've accidentaly mixed up lots calculation. When Martingale On, lot size is based on

risk per trade. but when martingale off, i used external lots setting (0.01).

I calculated lot size as follow:
1. Initial Lot is calculated such that when we loss (all losses in this system is 50pips), the amount of losses equals to

RiskPerTrade percent of account ballance. For example, Risk Per Trade = 0.2 (zero point two percent). Account Ballance =

$10.000,-
Then initial lots size is calculated so that if we loss, we only loss 0.2% X $10.000 = $20. So, lots size would be 0.04. of

course if we enable martingale, the risk per trade will become higher then 0.2%, depend on what level we are at.
2. If we enable Martingale, I provide 2 style:
a) style 1: for simplicity, based on our acc balance and risk per trade, assume initial lot size = 0.1. Max martingale level

= 4.
1st loss, lot size = 0.1 x (2 power 1 or 2 ^ 1) = 0.1 x 2 = 0.2 => level 1
2nd loss, lot size = 0.1 x (2 ^ 2) = 0.1 x 4 = 0.4 => level 2
3rd loss, lots size = 0.1 x (2 ^ 3) = 0.1 x 8 = 0.8 => level 3
4th loss, lot size = 0.1 x (2 ^ 4) = 0.16 => level 4
5th loss, lot size returned back to initial lot size = 0.1 because level = 5, which is is greater than max martyingale level
6th, 7th, 8th and so on, lot size = 0.1

b) style 2 same with above excep at 5th, 6th, 7th and so on, lot size equals to the last lot size, ie: 0.16
c) once we have a winner, lot size, in this case, initial lot, will be calculated again based on current account balance

about error when use lot = 0.01, maybe because your broker's minimum lot size is greater than 0.01. now i use

MarketInfo(Symbol(),MODE_MINLOT) to get info on broker's minimum lot size. on alpari uk demo acct, i found that min lot size

= 0.01.

inspired by your awesome web-based tester, i've added feature in the EA (updated on my previous post) to save the pattern to CSV file and then we can open it in excel and doing some pivoting to get summary about those pattern. you can choose number of chars of the pattern to be saved. to speed up saving process, disable TradeThePattern, so EA will not initiate do backtest, and only work to scan and record each pattern. conversely, if you want to only backtest, disable SavePattern2CSV.

attached below is an example of pivot table based on GPBUSD pattern.

stephen
Attached File(s)
File Type: xls 50pips candle.xls   243 KB | 567 downloads
  • #579
  • Dec 17, 2009 1:11am Dec 17, 2009 1:11am
Quoting maxwell-eqtn
Thanks C-speed.

I am glad you changed your mind and made an option such that we could experiment with different pip ranges.


Now the question is can some one come up with a rule for placing or choosing the orders such that wins and losses can alternate, but a win can not be two losses away in 12 or 14 trades. For example

L L L L L L L L L W L W
1 2 3 4 5 6 7 8 9 10 11 12 is a good win. As another example

L L W L L L W L L W L W
1 2 3 4 5 6 7 8 9 10 11 12 is also acceptable since a Win is not more than two wins away from a loss within...
Sorry, but i don't understand.
How can you win with 2 W and 10 L??
  • #580
  • Dec 17, 2009 6:46am Dec 17, 2009 6:46am
december updated :
initial method = 0 pips
evolution = 800 pips
2 traders viewing now