Experts: Arbitrage Triangle EURGBP-EURUSD-GBPUSD by Peter Mueller - page 3
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
So then what is the point of the EA? If it doesn't reflect reality and happens so rarely, why would you add it to your arsenal of trading tools, or was it just an exercise in creating an unrealistic example of something that will never happen, just for the sake of programming?
No offence intended, just wondering why it's necessary if it doesn't actually accomplish anything....
Arbitrage and correlation are certainly valid strategies, but if it's never going to happen in the real world because brokers have already stonewalled it, why teach people to stonewall?
Yes, as a training example of triangular arbitrage, this EA and the description on the MQL5 page look logical: the idea is to compare the synthetic rate with the market rate and open three trades in case of a discrepancy.
But it is usually very difficult to bring it tostable profitable trading on a real retail account, because the profit is "eaten up" by spread, commission and execution delays, and the discrepancy itself is often a fraction of a second. .
What exactly does the Expert Advisor do
It compares E U R U S D ( a s k ) / G B P U S D ( b i d ) EURUSD(ask)/GBPUSD(bid) with E U R G B P ( a s k ) EURGBP(ask ) and symmetrical condition via E U R U S D ( b i d ) / G B P U S D ( a s k ) EURUSD(bid)/GBPUSD(ask ) against E U R G B P ( b i d ) EURGBP(bid), and when the condition is met, opens three market deals with the same volume on EURUSD, GBPUSD and EURGBP.
In the MQL5 description, the same is presented as "automatic detection of arbitrage opportunities" and "opening and closing trades based on the calculated arbitrage potential". Closing trades is done by two functions (closing the "positive" and "negative" side), which simply go through all positions and close them by symbol and direction.
Why profit is not guaranteed
Discussions of triangle arbitrage in MQL5 explicitly mention two main problems: transaction costs (spread and commission) and execution delay, because of which the window of opportunity may disappear faster than all three legs open.
Even if the formula is correct (in MQL5 articles the logic is exactly the same: comparing cross rates and searching for an inequality for buying or selling), in real trading the price may "move away" between the first and the third trade, and then you get an unhedged position rather than arbitrage.
Additionally, it is important to take into account that some brokers have restrictions and sanctions against arbitrage practices in their documents, including arbitrage on delays and other "non-market" strategies, so profitability may be based not only on maths, but also on execution rules. .
What's in the code that prevents you from "squeezing" the result
The entry threshold takes into account "Total_Commission_for_Lot_Traded" as a value that is multiplied by the pip size, i.e. the commission (in money) is mixed with the price difference (in the quote), and because of this the entry filter may be incorrect..
Opening of three legs does not control that all three trades are actually executed (there is no check of the result of each trade and no logic of emergency hedging if one leg did not open or opened with slippage).
Closing positions does not filter by "magic number" and can actually close any positions by these symbols on the account, even if they are opened by hand or by another Expert Advisor.
Is it realistic to work out to profit
It is realistic to improve it to technically correct and safer execution (control of execution of each leg, filtering by "magic number", correct accounting of commissions and spreads, protection from partial execution and slippage), but the fact of profit will depend on the conditions of execution and costs, which are critical in triangle arbitrage.
If the goal is to make money on "pure" triangular arbitrage, it usually requires very fast execution and minimal costs, otherwise the divergence does not cover the cost of entry and exit.
So then what is the point of the EA? If it doesn't reflect reality and happens so rarely, why would you add it to your arsenal of trading tools, or was it just an exercise in creating an unrealistic example of something that will never happen, just for the sake of programming?
No offence intended, just wondering why it's necessary if it doesn't actually accomplish anything....
Arbitrage and correlation are certainly valid strategies, but if it's never going to happen in the real world because brokers have already stonewalled it, why teach people to stonewall?
I'll add a little from myself personally. CodeBase was not created to share profitable EAs with everyone. Programmers (including those from the MQL5 company) share their codes with all Forum members free of charge only so that you can take them as a basis and improve them at your own discretion. I think this is as clear as an axiom to everyone! I would like to emphasise that this is my personal opinion, not claiming to be the truth in the last instance ) .
Regards, Vladimir.
I'll add a little from myself personally. CodeBase was not created to share profitable EAs with everyone. Programmers (including those from the MQL5 company) share their codes with all Forum members free of charge only so that you can take them as a basis and improve them at your own discretion. I think this is as clear as an axiom to everyone! I would like to emphasise that this is my personal opinion, not claiming to be the truth in the last instance ) .
Regards, Vladimir.