Simple High Profit low Drawdown Triangular Arbitrage

@Everyone
Try this one on for size.

I got very busy the last time so I could not focus on this EA as I would like to, BUT I did manage to make some change.

The attached EA does not look at the virtual ASK and BID lines anymore.
Well it does for opening a trade, but not for closing it.

Instead it monitors the account equity until it is in profit and then only closes it.

I found that in a perfect world without massive slippage this EA will make a lot, but with slippage it is struggling to open at the correct place and close correctly too.
So now I only look at the actual profit.

This is still untested so be warned :0)

O and you can back-test with this EA 19.
I put a check into it for back-testing so that it disables the infinite loop.

My trades closed in loss due to swap commissions

@raffus
Unlikely.
Look in the logs at the slippage for the open and close of those trades.

The slippage is massive.
Swap is so small that it does not even play a role.

Also the EA does take into account commission.

Problem is slippage on the open causes the trade to be opened at the wrong position, thereby throwing the calculations out that is only looking at the virtual ASK and BID lines, which contains no slippage.

EA 19 I changed a few things in that it now checks for slippage after open and adapts the close calculations to that and also cater for some slippage during the close operation.
Take into account that EA 19 was developed at 2am with a coffee induces frenzy so ..... untested
I also suspect the profit take on 19 is too high, but I need further testing.

It did the test, loaded and waiting now

@Fxentropy
I suggest closing and reopening the chart window before all those line strangle you
I blame my cat, because he likes lines.
Otherwise maybe you like lines, in which case ... go for it
But you are talking about 20000 lines on screen, which might slow down a slow VPS, so less is better for speed.
Check your task manager or TOP on linux to see if your CPU is taxed, but unlikely.

I saw that the mt4 version creates some strange horizontal lines.
I want to fix it, but time management suggest I rather focus on implementing the strategy in the EA rather than play around with strings

The mt5 version does not seem to have that problem.

thanks for MAs



they look good for higher TFs

Right, done

we are very near to success excellent share going to use in live ECN account........
thanks

@badeel
Wow ok.
EA 19 is still untested.

I would strongly suggest not to use on live account.

I'm also working on some large changes to the EA, but it will take time.

Just saying.
Do not use on live.
A lot of work still need to be done to negate the slippage problem.

If you really want to throw money away then rather through it my way so I can pay for the development and spend more time on it.

Really this EA is not live ready yet.

If spread is a problem coucld your software work to get larger profit 20 or 200 pips?

Hi Jacques

I've been trading / playing with strategies for over 2 years now and your idea seems very interesting ???

Please send me some details regading you bot ???

email me at { email address deleted by staff }
we are neighbours i reside in jaohannesburg south africa.

Hi, jvbJaque.

I'm testing on demo, of course, the JVB Arbitrage Profit2 EA and Profit19.
Profit2 trades directly with three lots depending on your equity or balance (I'm not sure).
Profit19 draws lines on the GBP / USD chart, but does not trade. I do not know if it is looking for the right time to trade or is that it does not work on my platform.
What is the difference between them?

Thank you.

@Forex-Loser
Spread is actually not a problem.
Spread just determines how often an arbitrage opportunity it hit, but once hit then spread plays no role.

So spread determines number of trades and therefore also the amount of profit.

Increasing the profit take will not work seeing that arbitrage by its very nature does no deviate much from the ZERO line and therefore if you make the take profit too big then it will never hit it and never close the trades.

But it need to be big enough to cover the unknown slippage amount and that is what I build into EA 19.
EA 19 is just a rough test start and I will build on it to adapt to real world problems.



@maruishc
Hello neighbour
I must say I see your email address everywhere.
I even tried to send to your address {email address deleted by staff}, but it keeps coming back in that it cannot find the staff email address.
Must be lost at the post office

In any case. PM the address rather.

But what do you want to know.

It is not as if I am hiding anything about the EA.
Post #1 contain most of the information and you are welcome to ask anything you do not know or wonder about.
Like : Does it make coffe.
... Yes, but you need a special driver for it.


@artico01
Not sure I understand the question, but let me try and answer.
EA2 is VERY old and there has been a LOT of changes since then. Few thousand lines of code has gone through the EA and some of that new code have been discarded even in the later EAs.
So only use the latest EA for it contains all the updates.

EA 19 does trade, but is unstable.
Look at the top White line and Bottom Red lines.
They need to cross the opposite blue line before it opens any trades.
It also takes 100 minutes from the start of the EA before it start too look for entry positions.

The balance only comes into play once the EA tries to open a trade to determine the lot size.
After that it normally ignored the equity and balance in the older EAs and only looked the lines crossing the middle for close signals, but EA19 now looks at the percentage profit based on equity and balance for a close signal.
This was done because of slippage causing trades to be opened at the wrong positions and therefore closed at the wrong times.
Hopefully EA 19 or a later version will counter this slippage problem.

@Everyone

I love talking to everyone and try to answer everyone via forum, email, phone, whatsapp, pm,......
But I am overwelmed sometimes with the amount of interest in this.

Just wish my indigogo campaign had that amount of interest to fund this project, then I could have finished this already and not have gone broke

In any case.
Do not stop talking and sending, you are not bothering me, but if I missed your message then please remind me later.
I will not do it on purpose
Sometimes I miss stuff.

EA in progress. Spreads: UJ: 0.3 pips, GU: 0.2/0.3 pips. GJ: 0.9/1 pips. Time: 9:30 AM and 11:30 AM

ehee

Hello
what broker are you using?
It seems you have really good spreads

Thanks

Nice goal

@artico01
That is fantastic spreads.

What can I do to get an account like that ?

I'm still working on the slippage problem.

I found that slippage is extremely bad on MT4, but much better on MT5 so if you can use MT5 on that broker then I suggest rather use that.
I think it might be the way MT4 handles multiple trades or something.

But with an account like yours with those low spreads then I can increase the profit take a lot to cater for more slippage.
Still. I'd rather find a way to decrease the time it takes to open trades.
I also saw other low spread accounts with crazy high slippages/trade delays.
Does not help if the profit target is $100 and the slippage is -$150.

Also.
EA version 18 and older does not cater for slippage.
Rather use EA 19, which does extra calculations after opening a trade to counter some of the slippage.

I would like to see your logs if possible so I can improve the system, even if it is the logs of EA 18.
Just tell me what EA you are using.


@Fxentropy
That is very nice.
Is that on MT4 or MT5 ?
I'm still working on a massive update, but I ran out of time.
I hope to have it done by Tuesday.

Thanks to the "same lot bug" I actually discovered an interesting side-affect.
Yes if all lots are the same then the profit/loss fluctuates a lot more than balanced and calculated lots sizes.
BUT .... so far I have found that it still fluctuates around a ZERO point to some extend, just much more wildly and unstable than the balanced strategy.
The good thing about this is that is generates a large amount of more hits and potentially much bigger profits, at the expense of a bigger drawdown.
The drawdown can then be countered by decreased lot sizes and a small grid system.
Still a lot of tests to go, but who knows. Might be on to something.


@Everyone
I am still working on the slippage problem, which is caused by slow execution of trades.
More than 200ms per ticket some times, of which this strategy need to open 3 at a time, which causes massive slippage seeing that it takes up to 1.5 seconds to process 3 tickets at a time.

This seem to affect MT4 far more than MT5.

Maybe MT5 changes something, but even MT5 contains some slippage.

Any suggestion on how to execute 3 trade at the same time extremely fast will be appreciated.

I am looking into possible solutions, but so far I have not found a feasible pure metatrader solution yet and I started too look outside of metatrader, which unfortunately will exclude a lot of brokers that only supports metatrader.

The only way so far I found was to open a way out pending order.
Then modify that order once every 25 seconds to keep the trade server connection open, but that causes other problems such as overloading the trade server and angering the broker and getting your account blocked.

Other options are to use multiple threads, but I found that that actually makes things a bit worse seeing that now instead of one thread having to wait for a sign on then now you have 3 or 4 threads, so no improvement there.

This must be really bad for scalpers trading once every few minutes.
If the solution is out there then I will find it.
In the mean time there is wonderful JTrader and FIX and multiple other APIs.
Problem is that then the EA need to be adapted to fit each broker's unique API.
Not a problem, just takes time and I am working on a cross platform version of this EA to which you only add a small driver to connect it to your broker's API, but leave the base EA the same.
This is still far off in coding time of which I can only afford after hours midnight free development time.

Its MT4 Darwinex. Now running a second trade for today