Simple High Profit low Drawdown Triangular Arbitrage | Page 7 | 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

  • #121
  • Aug 18, 2017 11:41am Aug 18, 2017 11:41am
after zoom and modification looks like
Attached Image (click to enlarge)
Click to Enlarge Name: arbitrage image.PNG Size: 54 KB
  • #122
  • Edited 12:22pm Aug 18, 2017 12:11pm | Edited 12:22pm
Quoting jvbJacques
OK Here goes something. At last I gathered the funds to fix my server and along with it I made a massive amount of enhancements to the EA. Thing is. I did not have time to convert it to MT4 yet and with all of the enhancements most of the calculations are hard-coded into the EA so if you want to test it you need the following: Leverage of 200:1 ... Should work on other leverages, but I have not tested it long enough to ensure the draw-down and margin calculations are correct, but try it. Symbols GBPUSD, GBPJPY and USDJPY must be available for use....
ok just for helping you and save you time:

- never ever test this 3 currency hedge arbitrage with an demoaccount --> in demo account you will get always wins, thats stupid, really stupid. its like with all arbitrage models, you win in demo near always, in real accounts you dont.
- do it in real account in small or dont test it.

just to help you, you dont need to believe me. but demo trading is complete useless time, only testing your program is ok for prgramming errors, but not testing for any win.
  • #123
  • Aug 18, 2017 3:16pm Aug 18, 2017 3:16pm
Quoting badeel
after zoom and modification looks like {image}
@badeel
The modifications on the chart does not change the EA and it is still running fine as before.
It is just better visible for you now.

You might want to change the currency line colors to white so you can also see the currency line.


@dukas_trader
I know, but have got no funds to test on live account.
Need at least $1000 for proper test.
I'm struggling to get by month to month so that is not an option for me.
At least I am enjoying the math behind it and I have programmed a lot of AI programs for safety and security and the mining industries and multiple others.
And most of my programs have been taken from me by bigger companies for their own profit leaving me with scraps.
I'm a programmer. Not a lawyer or salesman and they know it.

So why not use what I know with what I enjoy doing, even if I ultimately can not benifit from it then at least others can.

Demo at least show some slippage problems.
I already found some brokers that seem to slip even demo accounts, which at least tell me to avoid them.

Demo is also more accurate than back testing.

The major difference between demo and live is slippage and one day I hope to test it.
I did build slip protection into the ea to hopefully take some slip hits.
  • #124
  • Aug 18, 2017 3:48pm Aug 18, 2017 3:48pm
Quoting jvbJacques
{quote} @badeel The modifications on the chart does not change the EA and it is still running fine as before. It is just better visible for you now. You might want to change the currency line colors to white so you can also see the currency line. @dukas_trader I know, but have got no funds to test on live account. Need at least $1000 for proper test. I'm struggling to get by month to month so that is not an option for me. At least I am enjoying the math behind it and I have programmed a lot of AI programs for safety and security and the mining industries...
i dont get this point.
you can test with very small amounts or even cent accounts in live trading. with higher amounts it get more difficult (not that it is not from beginning) because higer slippage, higher spread, not enough volume in all 3 currencies same time.....!

the difference between live and demo account are:
- slippage
- spread
- offquotes , unable to do trades to this prices for this amount
- to less amounts possible to trade
- different fill time, platform connections,responses ...
- sometimes different rollover fees
.....

you can program inside an ea what ever you want, until you test it in real account its complete wasted time, if you trade arbitrage trading, because arbitrage is in 99,99% of the time extrem wasted time, if you dont have special hardware/software advantage, its business of many people with all this, many competition for exact same trades.... and we dont speak about this cam software from all who offer arbitrage mt4 software!
  • #125
  • Aug 18, 2017 3:55pm Aug 18, 2017 3:55pm
Quoting dukas_trader
{quote} i dont get this point. you can test with very small amounts or even cent accounts in live trading. with higher amounts it get more difficult (not that it is not from beginning) because higer slippage, higher spread, not enough volume in all 3 currencies same time.....!

@dukas_trader
Very simple.
Read the theory in post 1.
USDJPY lot size is a calculation based on the lot size of Gbpjpy.

If the account size is too small then the lot size difference of USDJPY from the other 2 is nothing or too small to make a difference and to balance out the triangle.

You can trade with $100, but then it is more luck than calculations.

Best is $10,000 but $1000 will do
  • #126
  • Aug 18, 2017 4:25pm Aug 18, 2017 4:25pm
Quoting jvbJacques
{quote} @dukas_trader Very simple. Read the theory in post 1. USDJPY lot size is a calculation based on the lot size of Gbpjpy. If the account size is too small then the lot size difference of USDJPY from the other 2 is nothing or too small to make a difference and to balance out the triangle. You can trade with $100, but then it is more luck than calculations. Best is $10,000 but $1000 will do
and when you use a cent account?? a cent is in this accounts faked a dollar, so you have 100 times more in balance then its worth. but perfect for your situation.
so 100 usd is 10000 "account balance dollar" already to trade. this is not enough?

and so you get with small money much much more accurate results. demo will not help a little.
  • #127
  • Aug 18, 2017 4:37pm Aug 18, 2017 4:37pm
Quoting dukas_trader
{quote} and when you use a cent account?? so 100 usd is 10000 "account balance dollar" already to trade. this is not enough?
@dukas_trader
That actually might work.
What broker offers this with mt5?

Also what is the average spread you get with the 3 symbols on this kind of account?
What commission.
Sounds good.

As for the differences between demo and live and all the parts that can differ.
I see all of it as slippage.
The meaning of slippage to me is if you do not get the exact currency value that you opened on, whether it is because of currency changes, lag, spread or pink butterflies.
The point is that you picked a price point and did not get it.

Now I do realise there is always some slippage and I cater for it in my ea and then some, but some brokers do push it a bit.

My problem at the moment is lag, because I do not have a vps, but funds....

You might think $100 is cheap, but for some people like me that is a lot.
My cercumstances is of the sort that I live month to month and $10 here and there ads up for me.
I can not remember when last I had takeaway.

No coffee for me, unless I make it myself.
And I do love my coffee
  • #128
  • Aug 18, 2017 4:46pm Aug 18, 2017 4:46pm
Quoting jvbJacques
{quote} @dukas_trader That actually might work. What broker offers this with mt5? Also what is the average spread you get with the 3 symbols on this kind of account? What commission. Sounds good. As for the differences between demo and live and all the parts that can differ. I see all of it as slippage. The meaning of slippage to me is if you do not get the exact currency value that you opened on, whether it is because of currency changes, lag, spread or pink butterflies. The point is that you picked a price point and did not get it. Now I do realise...
slippage is only the best case you get, the worst case in real account is :
- offquotes: no trade possible because of connection, no quotes, or not this quotes for this amount you want trade
- order fill time is much slower...
- higher spreads
- much faster new pricequotes/refreshes

i dont use cent accounts, you have to look by your own to all brokers who offers this and check for all points you are interested in and make a table to compare, or open some with some brokers and take the best later..... (first demo to test ofcourse for some hours to filter what you dont like at all)
  • #129
  • Aug 18, 2017 5:41pm Aug 18, 2017 5:41pm
@dukas_trader

- offquotes: no trade possible because of connection, no quotes
This is bad, but an immediate close can solve this if all 3 symbols could not be traded.
- offquotes: or not this quotes for this amount you want trade
Slippage
- order fill time is much slower...
Slippage
- higher spreads than displayed when opened
Slippage
- much faster new pricequotes/refreshes/lag
Slippage

Slippage to me is a blanket term that states you did not get what you paid for for whatever reason.
You expect to get the amount at the time the order is placed, but I know there is some slippage.
Slippage can be broken up into multiple sub categories, but I do not care about that detail when the end result is the same.
The problem comes in when there is 20 to 50 pips slippage every time.
Once in a while will not hurt that much, but still a red flag against the broker if you have no lag on your VPS.

I checked a couple of cent accounts and although the idea is nice, their spreads are crazy high.
Spreads are important to this strategy, along with commission.
If I can get a cent account with spreads equal or better than MyFxChoice's ENC Pro account then that will be something to talk about.
MX looks nice, but I need a VPS to test it because on my server I get about 60pip slippage on basically every trade, which is far too big, but might be caused by my 280millisecond lag to them, where myfxchoice with a 250millisecond lag slips much less and therefore stays in profit, even with higher spreads.
  • #130
  • Edited 7:00pm Aug 18, 2017 6:14pm | Edited 7:00pm
Quoting jvbJacques
@dukas_trader - offquotes: no trade possible because of connection, no quotes This is bad, but an immediate close can solve this if all 3 symbols could not be traded. - offquotes: or not this quotes for this amount you want trade Slippage - order fill time is much slower... Slippage - higher spreads than displayed when opened Slippage - much faster new pricequotes/refreshes/lag Slippage Slippage to me is a blanket term that states you did not get what you paid for for whatever reason....
you play against big people with much money, big hardware and software.
arbitrage trading is what biggest do with biggest possible amount of competition.
many big firms win only from this (because most easy ,you dont need any trading plan or experience) and hide doing this by this. but the competition makes wins much smaller then with real trading, and for 99,99% of this trying even no wins at all.

its really only a extrem small chance that you can much from the table, if you have to trade with such small brokers.
try it, but dont put in to much hope.
arbitrage is fine in theory, but because its biggest competition, any tome advantage, any software advantage, any contact advantage, any hardware advantage from other kill you system in real trading. in demo trading you cant see this.
  • #131
  • Aug 19, 2017 10:53pm Aug 19, 2017 10:53pm
Interesting stuff u have, will test it on Icmarket demo. By the way, there see cents account broker that u can trade for live.
  • #132
  • Aug 21, 2017 5:27am Aug 21, 2017 5:27am
@EVERYONE

Bug detected.
The market start to move on Sundays for some symbols.
The EA then tries to trade when the market starts moving and fails seen that the market is not open for all the symbols yet.
This then messes with the calculations as partial and incorrect trades are then opened when the market opens and the EA crash.

Attached is a new EA that does not trade on Sundays and if trade failure is detected then close everything and try again.
Still looking for a better solution to detect if the market is open for all the symbols.
In MQ4 I can use the function MarketInfo, but this does not work in MQ5.
any help here would be appreciated.
Attached File(s)
File Type: ex5 2017-06-14 09h00 JVB Arbitrage Auto Profit12 FXChoice.ex5   129 KB | 451 downloads
  • #133
  • Aug 21, 2017 6:09am Aug 21, 2017 6:09am
Quoting jvbJacques
@EVERYONE Bug detected. The market start to move on Sundays for some symbols. The EA then tries to trade when the market starts moving and fails seen that the market is not open for all the symbols yet. This then messes with the calculations as partial and incorrect trades are then opened when the market opens and the EA crash. Attached is a new EA that does not trade on Sundays and if trade failure is detected then close everything and try again. Still looking for a better solution to detect if the market is open for all the symbols. In MQ4 I can...
simple check the last time for the 1 min timeframe. trade only when all 3 currencies have a time difference not more then 2 mins to actuell time or what ever time you want.
  • #134
  • Aug 21, 2017 6:26am Aug 21, 2017 6:26am
@dukas_trader
I was thinking the same thing, but I have read that there can be multiple reasons for the trading to stop or have gaps of multiple seconds or even a minute.
So this can be inaccurate.

But thanks.
I will build it in as another check.
The more the better, especially ones like these that does not influence the processing speed.
  • #135
  • Aug 22, 2017 8:28am Aug 22, 2017 8:28am
It seems like this strategy must have been tried by big players. Is there any indication that it is used by banks? My guess is that the market is not that simple.
  • #136
  • Aug 22, 2017 9:06am Aug 22, 2017 9:06am
@Kman
I also think the big player must be using this, especially the banks.
If you run the EA during back testing with the graph then ignore the profit and trades and zoom the chart out.
You will see that I draw 2 graphs and an average line between them.

The top graph shows what the actual account value is if you bought into the 3 symbols using the arbitrage calculation at the start of the run with and you do a SELL, BUY, BUY at that moment.
The bottom one is the same, but a BUY, SELL, SELL.

As you can see that no matter how long the period that the graphs are rock solid with minor fluctuations.
These fluctuations are what you can profit on, but they need to be big enough to not only cross each other, but to cover spread, commission and a bit of profit.
As you can see.
The big fluctuations are rare.

The advantage to the banks are that they have access you a fast price actions and EXTREMELY low spread with no commission.
What I would not do for spreads like that.
In stead of just making profits once in a while you can then make profits every few minutes or even seconds.

A good example is the broker XM.
They have extremely low spreads and crosses happen quite often.
Unfortunately their commission is very big and a cross need to be very big to cover the commission.
That and the fact that my poor server's internet connection is too slow and therefore receives a lot of slippage because of the execution speed.

If anyone can show me a broker with extremely low spreads and commission with very little slippage then please pm me, but then again without a good vps to that broker it probably will not help

I have been searching for a profitable forex strategy for many years now and have written hundreds of EAs based on advanced and simple strategies.
One thing I noticed is that the simple strategies seem to be more profitable than the advanced once in the long run.
This strategy is very difficult to implement in an EA, but the theory behind it is one of the simplest I found yet.

I have got other strategies that also show good profit and share one thing in common and that is they are extremely simple to understand.

Now some might disagree and I saw a lot of people saying this massive strategy and that is the best, but none can show proof of it and no one will share the detail of the strategy for others to test.
This is why in post 1 I shared everything.
I myself is dumbfounded that a strategy this simple can be so profitable.
Or at least I hope so.
I have got a massive amount of development ahead to test everything in the EA, but I am getting there slowly.
  • #137
  • Aug 22, 2017 10:12am Aug 22, 2017 10:12am
It seems like you would need a very large acount to make this worthwhile and the time to stare at your screen all day.
  • #138
  • Aug 22, 2017 10:38am Aug 22, 2017 10:38am
@Kman
Surprisingly no to both.

Test my EA with backtesting to see the results.

Account size.
I found that it can be profitable even on a $100 account, but I would not suggest that, because at that account level the calculations do not have enough room to work and therefore it is VERY risky and has the potential to not be a good hedge and blow your account, even though I could not do that during back testing it did come close a couple of times.

I'd say the minimum account size with a 1:200 leverage is $500.
Good = $1000
Best = $10000 or better.

As for time staring at the screen.
That is what I am teaching my EA to do for me.
The big trades are few and happen fast at any time (normally while you are on the toilet), but the EA can pick them out and you do not have to stay awake at all.

I still have a long way to go, but it does look extremely good with about 1000% profit in 6 months based on "real ticks" and crazy amounts going back further, but that is more to do with the inaccuracy of the data I use in that I only have real tick data dating back for about 1 year and no more.

I only now started forward testing, but my server is a bit slow (400 millisecond delay) to really do justice to the trades, but it hit 3 very good opportunities last week since Tuesday.
  • #139
  • Edited 11:31am Aug 22, 2017 11:09am | Edited 11:31am
Quoting jvbJacques
@Kman Surprisingly no to both. Test my EA with backtesting to see the results. Account size. I found that it can be profitable even on a $100 account, but I would not suggest that, because at that account level the calculations do not have enough room to work and therefore it is VERY risky and has the potential to not be a good hedge and blow your account, even though I could not do that during back testing it did come close a couple of times. I'd say the minimum account size with a 1:200 leverage is $500. Good = $1000 Best = $10000 or better. As...
you still count demo wins, so you count nothing. dont even think about this win will happen in real account until you used real account. you really cant invest time in demotrade with arbitrage systems, its 100% wasted time until you test it in real accounts.

some strategies you can really count in demo account and compensate against real account with some pips or some intern recalculation, but with arbitrage really all you do in demo account is like you do nothing at all. any result is 100% useless and not this like in reality.

in all your time, you already invested in this demo trading part, you could invent a normal trading model what you can start with very less money.
i know, reality is hard, but basic in trading forex you cant go around, thats why no goo dtrader would any other beginner say "hey start with demo account". demo account is for lets know the platform and for test programs for coding problems. but for testing trading always you need a good real account, even its very very small its 1000% better then any demo account can even be. you learn also to grwo slowly and dont be greedy.

i know you dont want to hear this, i dont will say it you again. but you cant lose so much time with worthless things. first learn basics whats possible, whats not and how its possible to do same thing, then doing. wasting time is not best way, because time is often really money in this business.
  • #140
  • Aug 22, 2017 11:30am Aug 22, 2017 11:30am
@dukas_trader
I have got a different view of things.

You seem to think that if you do not try then you can not lose.
I'm more of the thinking of that if you do not try then you can not win.

I can not go to a live account without first developing the strategy and testing it using backtesting, then demo then live.
If I jumped to live using the first iteration of the EA then I would have lost everything.
Not because of the strategy, but because of EA bugs and the wrong implementation of the strategy.

It is easy to talk about the strategy, but difficult to tell a computer EA to do what you think.

I count nothing as wins, but I do see the potential and potential is all I see until I use a real account at the end of the development.
I know monopoly money is worth nothing in the bank, but it is fantastic to test with.

If you build a racing car for the first time in your life you do not go to a live race.
You will break the car, crash and die or kill someone.
You first go to a demo track with the perfect conditions and ensure everything is working there.
Then you start to through rocks at it and simulate problems to break it so you can fix it and make it better.
Then after some time of development you go to a small race a couple of times and do the last small adjustments before going to the big live account.

I think demos are fantastic for testing.
No need to kill an idea before it can be fully realized.
An if it does not work in a real account as you say then so be it.

If we stopped development every time something is difficult then we probably would still live in caves or worse.

It costs you nothing and I'm going bankrupt while doing this(Not because of this).
So if it work then good for everyone and if not then I'm the only one that risked anything.
But if it works then I want everyone to benefit from it.
0 traders viewing now