dsugums 6 pip auto-trader by 'im again

[Jeuro]

Raiden is correct. I hope everybody understand.

By using same lot size for all 3 pairs it is an "imperfect hedge”, therefore doing it is just making a small buy or sell EUR/CHF. (depending if doing buy/buy/sell or sell/sell/buy). the 6 pips profit would come from the direction the EUR/CHF takes.

It is possible to make a “perfect hedge” (only mathematically, because most brks do not have 5 or more digit lot size.) In the event someone is using a brkr with lot size of many decimals, then and only then doing the perfect hedge has no risk. (only spreads) But it is impossible to make money without risk, so no point in doing that.

The “perfect hedge” does have a bit “market” adjustment that runs between -1 to +1 (equation EUR/USD*USD/CHF=EUR/CHF ) but impossible to take advantage of it. And that is a different subject.

Trading what I call “equations” is not a bad idea. I mean, it is almost the only thing we “sort of” know the future. EUR/CHF will always be EUR/USD*USD/CHF (many others equations exist ). And that is an excellent starting point to create systems with very calculated risk. But back-testing multi-pairs is always the ultimate challenge to speed process from theory to facts.

.

[bazze]

Hi...
I'm a newbie to forex,and I wanna say a big THANK U to all of you on this forum who let us try out your ea's ....I'm sick and tired of $97 trash

[wickedwin]

Works until we get a central bank intervention and the pair correlation is completely destroyed. Meaning this strategy will blow up your account eventually.

[forexglee]

Trading risk free is a beautiful thing.

Wish it were possible.

I hate to be the "wet blanket" here but retail forex arbritage probably cannot be automated successfully with the tools we currently have for several reasons:

1. ONLY the big houses with realtime "floor" access (probably the market makers themselves) have a chance to make any consistent money on arbs. I have studied this extensively. It is useful in learning about trends, but you can't trade it profitably, especially when you limit yourself to a single crim on a Triangle Arb. Hanover correctly pinpointed the hurdles that must be overcome with Triangular Arbitrage on the same broker.

2. The slippage from order execution alone will kill this, not to mention the fact that there is no guarantee of identical transaction time completion, even with market orders across brokers, which logically would offer the best chance for success.

3. Setting up arb trading across brokers might work, but not with the Empty4 for reasons clearly laid out by MBKennel earlier. We would need a FIX account and a platform that could execute on 2 or 3 brokers on what effectively should be the same tick. Even with an ECN commission of say 0.8 it will be dicey.

MBKennel hit the nail right between the eyes when he said "Sorry to be a party-pooper, but exploiting transient anomalies in pricing is not something that we can do, when there are Very Powerful Gorillas With Lots of Money Who Do This For a Living".

However, it seems to me that if we set up a Currenex or similar account to compare their prices against selected retail crims who give decent, timely fills there may be a chance to see this work. if the crims didn't figure out what we are up to first. This probably would not happen as each crim account would be traded individually, so one or two accounts would accumulate profits while the other would be negative.

We have DDE capability to fill an excel spreadsheet where can see the price differences between the retail crims now, but would need software to compare prices between the market maker and the downstream crims in real time. Maybe we have someone here to code this interface so we can study the feeds on the different brokers, but it has to be live feeds not demo.

Arbs between the market makers and the downstream retail crims happen all the time, but in increments of 0.5 - 1 pips. Most crims broadcast "composite" feeds they make from 2 or 3 market makers. This is where the arbs pop up. We might find a "glory hole" of arbitration between a market maker and a few crims after careful research. If the research pans out then we can we "industrialize" this endeavor.

The 5 or 6 pippers do happen, but infrequently in comparison. When they do, they unfortunately come with risk, whereas the .05-1 pip, perhaps 1.5 pips are probably popping up to the tune of a few thousand per day between the market makers and the crims.

Don't know if this helps. Might be fun to give it a go! Nothing ventured, nothing gained.

Cheers!

George

[james_fx]

Hi Steve

Thanks for coding the EA in lightning speed. Told you it can be done in a jiffy

OK, the EA CANNOT be backtested, so can't do any curve testing. So, only way to do it forward testting. I have done it manually but the 5-6pips only appears few seconds! My fingers not fast enough to close the trades using script.

EU, UCHF and ECHF are hedged one another. Once opened, the trades will be moving in twilight zone and suddenly, boomed, you get 5-6pips (or $) in profit for few seconds. Grab it and run the trades again. I will not opened any new trades after US closes because the spreads increases but eventually will normalised.

Only EA can do this trades. Let it run on demo and see how the trades behave. Other combinations includes:

EU, EJ, UJ

Try out other combs as well.

Thanks Steve and Dsugums for this new toy
As mentioned by Dsugum about the not opening trades after US close,
maybe a parameter to stop trade opening during certain hours can be added?

Thanks again guys

[SteveHopwood]

Latest update in post 1. I coded the EA quickly yesterday and forgot to convert some of the code to multi-pair use.

Also, I had an idea based on T101/ES Trader ideas. I describe this in 'Stacking' in post 1.

[mobthehop]

Steve - there are now two versions in inital post - suggest that only the latest version should be avail for d/l....

Would it make sense to refers the three pair set in terms of buy / sell, give it a different magic #, attach it to a different chart?

Cheers

[SteveHopwood]

Steve - there are now two versions in inital post - suggest that only the latest version should be avail for d/l....

Would it make sense to refers the three pair set in terms of buy / sell, give it a different magic #, attach it to a different chart?

Cheers

Thanks. I forgot to delete the old one. That had bugs in it and needs replacing with the latest, regardless of whether people are going to play with stacking.

Cheers

[Jeuro]

[quote="forexglee"]Trading risk free is a beautiful thing.

Wish it were possible.

I hate to be the "wet blanket" here but retail forex arbritage probably cannot be automated successfully with the tools we currently have for several reasons:
-------------------------------------------------
Hello George.
I am not sure if the original proposed idea was to do a triangular arbitrage. I mean from the fact that it proposes to do equal lots in all 3 pairs. Doing same lot size in all 3 pairs is just an imperfect way to hedge and the results, profit or loss depends if one took the right (buy/sell) Eur/Chf. Basically trading the EUR/CHF directional and wait until in profits ( and that could be forever while getting big dd)

But your are completely correct. No one has been able to automatize and make money doing a triangular arbitrage ( using the perfect lot sizes)


.

[garyfritz]

Hi Gary, are your calculations of profit taking the three spreads into account?

No, I'm just looking at the bid. Maybe that's where I'm missing something. But it looks like several experts on triangle hedges &etc have expressed doubt about this strategy.