Simple High Profit low Drawdown Triangular Arbitrage | Page 3 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Triangular Arbitrage 279 replies

Triangular Arbitrage 16 replies

Triangular Arbitrage and carry trade 6 replies

Triangular Arbitrage 7 replies

Question: Trade Mechanics of Triangular Arbitrage 6 replies

Attachments: Simple High Profit low Drawdown Triangular Arbitrage
Exit Attachments

Simple High Profit low Drawdown Triangular Arbitrage

  • #41
  • Jun 17, 2017 3:23pm Jun 17, 2017 3:23pm
@kosmo
I understand. thanks.
  • #42
  • Jun 17, 2017 6:40pm Jun 17, 2017 6:40pm
Quoting tunera
let me tell you, the only way to have any sort of profit or loss during a triangular arbitrage is when the lotsizes of the 3 trades are not perfectly calculated (you are more exposed on some currencies than the other), thus your profits/losses comes from the fluctuation of such currencs. Basically you are buying and selling randomly and taking many small profits, what can ever go wrong?
then open only the difference, simple math, n dont pay truck load of spread comission slippage swap for rest of the entries
that will take u single entry conventional trading
WaBaS Where are Buyers and Sellers
  • #43
  • Edited 2:02pm Jun 18, 2017 2:45am | Edited 2:02pm
I'm subscribing this thread. this is something i want to try out in and demo account. this strategy can be useful.
there is a broker named TradersWay offer 1:1000 leverage. i don't work or have any affiliation with them. I only found the info from forex broker reviewers and doing google searching. Also got a account with them as a regular ecn account.

hopefully this EA can work on smaller accounts and somehow try to upscale the account balance equity by doing smaller lots or so.
  • #44
  • Jun 18, 2017 4:28am Jun 18, 2017 4:28am
@mr.brown
simple math yes ...
Starting to get complicated to build it into an EA, but still a lot of fun

@rogelio11
Thanks for the broker roge.
Keep in mind I build this EA just as a test and testing exclusively using a broker called myfxchoice using a leverage of 200:1.
Anything else might mess with its head.
I plan to put a lot more work into it to make it compatible with any flavour of broker settings out there, but it will take time.
But still ... 1000:1 looks very good.
I think I'm going to test it a bit.

My other big problem is metatrader 4.
Yes I did convert it to 4, but I have got no way of testing it, because mt4 can not test multiple currencies at once, so problems can creep in.
Honestly I actually prefer MT4.

For everyone.
Keep in mind the EA is unstable.
I already tested it on another broker and it failed to pick up some settings like swap for some reason.
If you discover other problems let me know.

As for the theory.
It still hold firm and true.
I just need the EA to be smart enough to cater for the technical problems and feed it into the theory.

Almost time for me to go back to work so the updates might slow down until next weekend.
Im trying to find alternatives to help me continue on this and will let you know as things progress.
  • #45
  • Jun 18, 2017 4:42am Jun 18, 2017 4:42am
Ok this is risky, but I need advice.
I opened an indiegogo page to help me fund some of the development of this EA seeing that I am received requests and PMs from everywhere for a lot of setting changes and enhancements.
I appreciate it, but it is a lot of work.
I wont paste the link here just yet seeing that I do not know if it is allowed here.
That is my question.
Can I or can't I post a link ?
This EA will be free once done and even before as you can see now in any case, but I need some way of helping or just to speed up the development.
  • #46
  • Jun 18, 2017 5:51am Jun 18, 2017 5:51am
Quoting jvbJacques
Ok this is risky, but I need advice. I opened an indiegogo page to help me fund some of the development of this EA seeing that I am received requests and PMs from everywhere for a lot of setting changes and enhancements. I appreciate it, but it is a lot of work. I wont paste the link here just yet seeing that I do not know if it is allowed here. That is my question. Can I or can't I post a link ? This EA will be free once done and even before as you can see now in any case, but I need some way of helping or just to speed up the development....
Jacques, I suggest you message twee and check with him. Any link leading towards fundraising will/may turn your thread in to commercial.
  • #47
  • Edited 6:29am Jun 18, 2017 5:56am | Edited 6:29am
@shiva
Thanks shiva. I will rather then not include any links.
My goal is in any-case for everyone to benefit from this and I must say I have seen a lot of shortcomings in the EA which I must look into thanks to this thread.
Still. The theory remains unchanged, which is a good thing as it shows this is working as expected. So far.

Wow to spreads
I see now what you mean about so few brokers offering spreads in range of the EA maximum spread limits.
I have been using fxchoice's enc account and these spreads are quite normal for them, probably making up some of the difference using their commission, which I really do not mind paying anymore seeing the massive spreads of other brokers, even other enc commission brokers.

This is something I need to look into and I will.
The program need to take into account the different spreads and cater for it in the profit, without moving the ZERO point down, which is happening at the moment.
Using the myfxchoice spread the ZERO point does not move down much and therefore no need to adjust the EA for it.
But on other brokers the movement is bigger and sometime so big that the amount of profit that must be made is more than the entry point.
I should be able to cater for this and have opened an account with fxpro enc, which have got a crazy big spread compared to fxchoice, but a lower commission.
I will let you know when I have adapted the EA for this seeing that if it can work between such different brokers then it should also work with a range of others.

As for making the spread a user parameter.
That should not be necessary, because the coding of the program can pick it up.
At the moment there is no coding to cater for high spread so it just ignores it and just do not open a trade, but in future I will find a way to run even on high spreads by taking into account other factors like commission in the final profit, thereby allowing for higher spreads.
Coding wise half of the commission is only added to the position once you close it, where spread in a sense is already mostly added once you open it, and the program need to be taught this so it does not calculate the ending profit incorrectly.
Sorry. Lots of techo babble.

Point is keep an eye on post #1.
When an update comes I will update post #1.

One thing to keep in mind with this strategy is that it is sensitive to spreads.
It opens large lot positions and holds it for extremely small profit/loss targets.
Therefore big spreads will hurt it a lot.
Commission on the other hand does not affect it so much.
So I rather go for low spread/high commission than the other way around.


@EVERYONE
Stupid question.
I made a mistake and opened probably the wrong account on fxpro, but this is a good thing seeing that it revealed something I need to know.
Comparing my fxchoice to fxpro accounts.
fxchoice swap=about $30 over 2 years for a very small lot size, but $3 of commission.
fxpro swap=about $754 over 2 years for ZERO commission.
Poor EA was crying after this test so I need to teach it not to panic and cater for it.
Can swap differ so much between brokers?
I though it was a currency connected value, but it seem now to be broker/account specific.
Am I correct ?
  • #48
  • Jun 18, 2017 6:33am Jun 18, 2017 6:33am
Quoting jvbJacques
Poor EA was crying after this test so I need to teach it not to panic and cater for it. Can swap differ so much between brokers? I though it was a currency connected value, but it seem now to be broker/account specific. Am I correct ?
How do you backtest on MT5?
  • #49
  • Jun 18, 2017 8:23am Jun 18, 2017 8:23am
Quoting jvbJacques
******************************* Very basic summary ******************************* 1) Buy 1 unit GBPJPY 2) Sell 1 unit GBPUSD 3) Sell ?? units using the USD in point (2) in USDJPY to balance everything out Do this randomly at any time if no open positions. Close in profit and open other direction (1) Sell (2) Buy (3) Buy Close in profit. Rinse and repeat. Can this work or what is the pitfalls with this idea ? ******************************* Detail ******************************* OK I am still new in this, but must say triangular arbitrage looks...


Hi,

If I buy 1 unit GBPJPY and sell 1 unit GBPUSD then how many units of USDJPY do I sell? Sorry for asking but I don't understand the "??" in para (3). Why can it not be used on a live account?
  • #50
  • Jun 18, 2017 8:38am Jun 18, 2017 8:38am
@shiva
Attached is a picture of the tester.
Open metatrader-->view-->strategy tester
Pic the options the same as in the bottom tester part of the picture or make adjustments as you require.
I also attached an EA I made to show the affect of triangular arbitrage.
Basically it just opens up a 3 pair trade and never close it.
This way you can see how it stays at the ZERO point, except for the swap that will bring it down a bit, depending on your broker.
As stated myfxchoice have got a low swap where fxpro is high.
Other brokers will also be different.
Once the test start click on graph at the bottom to see you profit loss run up and down.
Another window will also open up.
At the top of this new window is pause and stop button and next to it a slider which you must pull all the way right to increase the test speed.
Beware.
Install and use the metatrader supplied by the broker you want to test seeing that I found sometimes that if you use the same metatrader between different brokers that the tester will use the settings for the owner broker on the others, which will give you incorrect results.

@rajandaas
GBPJPY == 1 lot
GBPUSD == 1 lot
USDJPY == 1 lot x GBPUSD(1.27778 lots)
Think of the EA as a very successful race car that you know will win everything according to the specs on paper.
And even during testing the race car is the best.
BUT the engine parts of the race car is put together using duct tape and its weal nuts aren't tied down.
So sooner or later it will fall apart.
This was done so that I can quickly build the race car as a proof of concept and to test the theory, but I still have a LOT of work to remove the duct tape and properly fasten everything so that it can run on any broker without falling apart.
Things like internet problems or metatraders crashes need to be catered for in the EA, which is not at the moment and the EA can not recover from it at the moment.
Things a test run will not show, but I know about it and will insert it as I get time and funds in to do it.
I will keep post #1 updated with updates and when it becomes stable enough for live environment I will also post it there for everyone to use.

Not only this, but this EA is very basic and I have already seen multiple ways to increase the profit even further and decrease the drawdown more, which is not implemented yet seeing that I do not want to over complicate the EA at this stage until my original theory is running stable.

So DO NOT USE ON LIVE ACCOUNT PLEASE. NOT YET
Attached Image (click to enlarge)
Click to Enlarge Name: Screenshot from 2017-06-18 15-19-57.png Size: 118 KB
Attached File(s)
File Type: ex5 2017-06-12 09h00 JVB Arbitrage Auto Profit3 Single Trade Showing off triangular arbitrage.ex5   90 KB | 636 downloads
  • #51
  • Jun 18, 2017 9:20am Jun 18, 2017 9:20am
Quoting jvbJacques
@shiva Can swap differ so much between brokers? I though it was a currency connected value, but it seem now to be broker/account specific. Am I correct ?
Yes it can. Pairs involving currencies like Aud and Nzd will cost the trader more across all brokers due to the respective bank rates, but not all brokers pass that through the same.
  • #52
  • Jun 18, 2017 9:23am Jun 18, 2017 9:23am
Jacques

How do you calculate the base lot size. Lots per x$ of balance?
  • #53
  • Jun 18, 2017 9:40am Jun 18, 2017 9:40am
@spyderman
Thanks.
I will adapt my EA to detect swap and compensate.
Should not be too difficult seeing swap make for an EXTREMELY small part of the profit.
Just be aware that the current EA does not cater for high swaps and will DIE.

As for the base lot.
That is a bit more involved.
I calculate the margin required per lot size and then how how many levels deep I want to go, which at the moment is 2 level, but I am thinking of increasing it.
Then calculate the amount of profit needed per trade using the amount of lots based on margin requirements and then using a simple loop decrease or increase the amounts to fit the 75% of the account balance.
Once this is done I use the gap between the (margin for all lots / amount of levels) and the account balance as the stop loss level, which normally account for 25% of the account, but could be less.
Then I use the difference of the equity before and after opening the trades and add $0.5 per 0.01 lot used for one symbol as the take profit point and also the next level down grid point.

This is not all there is, but the rest of the details are very boring.
Point is that I cater for failures in the calculation and try to use the whole account balance, while taking in account the maximum loss and profit and margin used.

Now that was a mouth full wasn't it
I just love calculations.

The theory might be basic, but implementing it not.
But then again that is part of my nerd powers .... That and star trek reruns
  • #54
  • Jun 18, 2017 9:55am Jun 18, 2017 9:55am
Sterkte met jou thread Jacques!

Subscribed!!!
Ask and you look stupid for 5 minutes. Don't ask and stay stupid forever.
  • #55
  • Jun 18, 2017 9:59am Jun 18, 2017 9:59am
Quoting jvbJacques
@spyderman Thanks. I will adapt my EA to detect swap and compensate. Should not be too difficult seeing swap make for an EXTREMELY small part of the profit. Just be aware that the current EA does not cater for high swaps and will DIE. As for the base lot. That is a bit more involved. I calculate the margin required per lot size and then how how many levels deep I want to go, which at the moment is 2 level, but I am thinking of increasing it. Then calculate the amount of profit needed per trade using the amount of lots based on margin requirements...

Inserted Video
  • #56
  • Jun 18, 2017 10:02am Jun 18, 2017 10:02am
@EeZeeTrader
Wel Hallo daar.
Nogals lekker om 'n ander Suid Afrikaner hier te sien

@dagoods
I agree.
The answer for anything complicated and boring
  • #57
  • Jun 18, 2017 10:06am Jun 18, 2017 10:06am
Quoting jvbJacques
What broker do you suggest that is trustworthy with 500:1 and 0.001 lot size ? That will still only take it down from $10,000 to $1000, but still a massive improvement.
I'm using FXPro and have to say the service is good. Money deposits and withdrawals is quick and doesn't cost an arm and a leg.

ACM Gold is locally based in Sandton JHB but their spreads are quite high but also gives you a 500:1 leverage.

Regards

EZ
Ask and you look stupid for 5 minutes. Don't ask and stay stupid forever.
  • #58
  • Jun 18, 2017 10:18am Jun 18, 2017 10:18am
@EeZeeTrader
For some reason is there no commission and very high spreads and very high swaps on the demo account I opened at FXPRO.
This is compared to my fxchoice accounts.

According to their website they do charge commission and have low spreads and swaps so I am confident if I take more time to find that demo account it will do better.
I probably did something wrong and will check into it a bit later, but for now I am using it as a worst case scenario.

I like to make sure my programs can work with it so that the better accounts can make even more profit.

ACM Gold.... High spreads will hurt this strategy seeing that you open big lots and take very little profit.
Therefore high spreads goes hand in hand with spreads.
Meaning you start lower than normal and need to make more profit than other brokers.

The very nature of triangular arbitrage is the small up and down movement and therefore if you start too low then you might never reach the profit point, seeing that unlike normal forex trading this have got a top and bottom level that changes very rarely and if so then very little so NONO to spreads.
Rather go for commission and low spreads.
  • #59
  • Jun 18, 2017 10:59am Jun 18, 2017 10:59am
Quoting jvbJacques
@EeZeeTrader For some reason is there no commission and very high spreads and very high swaps on the demo account I opened at FXPRO. This is compared to my fxchoice accounts. According to their website they do charge commission and have low spreads and swaps so I am confident if I take more time to find that demo account it will do better. I probably did something wrong and will check into it a bit later, but for now I am using it as a worst case scenario. I like to make sure my programs can work with it so that the better accounts can make even...
What if you open a cashback/rebate account where you get back a percentage of the spread returned to you?
  • #60
  • Jun 18, 2017 11:18am Jun 18, 2017 11:18am
@shiva
Where can I find more information on these rebates ?

Remember.
My expertise lie in programming and mathematics.
Not in forex.
This strategy is just perfect for the math side of me and the EA for the programming side.

OK I do have a few years of forex back ground, but that was more just dabbling in it until now.

I do not know the inner workings of brokers.
0 traders viewing now