My Own Broker Arbitrage | Page 9 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Arbitrage with no-interest broker 147 replies

eur/usd arbitrage broker? 4 replies

does you broker ban arbitrage on manipulated price quotes? 2 replies

My idea for broker arbitrage 13 replies

Broker arbitrage BIG TIME 4 replies

Attachments: My Own Broker Arbitrage
Exit Attachments

My Own Broker Arbitrage

  • #161
  • Dec 30, 2011 4:25am Dec 30, 2011 4:25am
Quoting nondisclosure00
Now this throws a monkey wrench into what metaquotes told me.
I'll see if i can think of a way to demonstrate this without setting up an overly complex test - i guess it should demonstrate a blocking OS call - while using the Win32 API to access the current thread. AT least we could then get the whole picture - I could easily be mistaken as i was chasing effect rather than cause when i implemented the system

I'll see if i can do that later today
  • #162
  • Dec 30, 2011 11:10am Dec 30, 2011 11:10am
Thank you a LOT ianj1! Let me know what you find.

Quoting ianj1
I'll see if i can think of a way to demonstrate this without setting up an overly complex test - i guess it should demonstrate a blocking OS call - while using the Win32 API to access the current thread. AT least we could then get the whole picture - I could easily be mistaken as i was chasing effect rather than cause when i implemented the system

I'll see if i can do that later today
  • #163
  • Dec 30, 2011 11:12am Dec 30, 2011 11:12am
Quoting Ronald Raygun
What I find most bothersome is how some people don't read the instructions they are given.

To those who whine that it won't work, build the system and prove me wrong. Stop speculating. I'm here busting my ass trying to prove to myself whether or not it will work, and when it's done I will release it
[center]for free.

[left]Your collective discouraging comments are not constructive to this thread. It's like I tell my staff, "Don't come to me with problems, bring solutions too."

Further:

To...

Thank you Ronald for all your hard working and i really wish that all the members will help and share with their comments, so we can all have something good for ourselves!! By the way the Scripts are working well now but it needs some Inputs like Number of Lots and also to make it accept pairs as "EURUSD. / EURUSDm..." i am trying to find a solution for this issue and i will let you know once i have it .. Thank you again!!
  • #164
  • Dec 31, 2011 9:15am Dec 31, 2011 9:15am
Hello,

Ronald, you're a genius and everybody appreciates your hard work and the fact, that you're releasing it for free is even more amazing. But arbitrage will bring you all trouble or even get your positions cancelled/account banned. Brokerages know what is going on, trust me. I can tell that from my own experience.

Regards
  • #165
  • Dec 31, 2011 10:10am Dec 31, 2011 10:10am
Quoting nqm0t0
Hello,

Ronald, you're a genius and everybody appreciates your hard work and the fact, that you're releasing it for free is even more amazing. But arbitrage will bring you all trouble or even get your positions cancelled/account banned. Brokerages know what is going on, trust me. I can tell that from my own experience.

Regards
Then what you're doing isn't true arbitrage.
  • #166
  • Dec 31, 2011 11:15am Dec 31, 2011 11:15am
Is that because "true" arbs don't always play out and will have some losses? An account making 1,995 profitable trades out of 2,000 will naturally raise red flags, but one with an 80/20 win loss ratio would stealth the system as just another EA based on technical analysis and is in no way malicious?



Quoting Ronald Raygun
Then what you're doing isn't true arbitrage.
Skype: heliosphan187
  • #167
  • Dec 31, 2011 2:47pm Dec 31, 2011 2:47pm
To a lot of folks in the forex world, arbitrage is using a fast broker's feed to predict a slower broker's feed.

That's not what I'm targeting.

Arbitrage requires buying in one market, and selling in another simultaneously. In this case, we are looking for two brokers with two different prices for the same pair. If the buying price at one broker is less than the selling price at another broker, then if you buy from the buying broker and sell at the selling broker, you turn a profit.

Given the nature of retail forex, you cannot simply buy a pair with one broker, transfer it to another broker and sell it. Instead, you go long with one broker and short with another broker simultaneously. In doing so, you expect the price to move to the middle. At that point, you close the trades with a net profit.

The benefit of this sort of arbitrage over time-lag arb is that I can in theory hold the trades open indefinitely (or whenever an account runs out of margin. This means that while I can enter a trade where the spread (bid - ask) is negative, I can also wait for the spread to widen before taking my profits.
  • #168
  • Dec 31, 2011 3:52pm Dec 31, 2011 3:52pm
Quoting Ronald Raygun
To a lot of folks in the forex world, arbitrage is using a fast broker's feed to predict a slower broker's feed.

That's not what I'm targeting.

Arbitrage requires buying in one market, and selling in another simultaneously. In this case, we are looking for two brokers with two different prices for the same pair. If the buying price at one broker is less than the selling price at another broker, then if you buy from the buying broker and sell at the selling broker, you turn a profit.

Given the nature of retail forex, you cannot simply...
RR,
From what i have seen , the quoted price difference (atleast with retail bucketshop brokers) is almost always caused due to one of the brokers running a slow feed. Or their prices not refreshing or reacting fast enough during news events. This is normally tagged as lag trading with these brokers and more than likely causes them to cancel the profits. I had a pretty decent arb system developed about 4-5 years ago that i used for a little while on some of these brokers until they cancelled out my trades.
I am pretty sure you will be running this on ECNs or higher tier brokers where i have no experiece with... howeever if anyone is planning on runiing this on the bucketshops of the world, iam pretty certain it wont end good.

Aside from the actual system development that you have been very diligently doing , we also need to identify the broker systems where this will work. Hopefully once the system development is completed(no pressure), we'll start identifying brokers (offline if preferable) where this will work.

Regards
  • #169
  • Edited 5:08pm Dec 31, 2011 4:00pm | Edited 5:08pm
Quoting andieliu
still run at demo, sometimes master script must be reloaded or drag to chart again, because stopped/hangs
today start
Nice work RR thanks for sharing so soon!

yours looks good mine has open trades still ????
think I may have inputs wrong on master

can you please tell me how you set up your inputs for the master

thanks in advance
  • #170
  • Dec 31, 2011 7:12pm Dec 31, 2011 7:12pm
Quoting bravester
RR,
From what i have seen , the quoted price difference (atleast with retail bucketshop brokers) is almost always caused due to one of the brokers running a slow feed. Or their prices not refreshing or reacting fast enough during news events. This is normally tagged as lag trading with these brokers and more than likely causes them to cancel the profits. I had a pretty decent arb system developed about 4-5 years ago that i used for a little while on some of these brokers until they cancelled out my trades.
I am pretty sure you will be running...
A broker's liquidity providers' liquidity providers ultimately are interbanks, and they are sometimes overexposed / underexposed to various currencies which causes a run or a rally on that pair. Now because each interbank does not have the same amount of exposure to each currency pair, you can expect each of them to quote different prices, and some instances where the bid from one bank is greater than the ask from another bank. That's the ideal opportunity.
  • #171
  • Dec 31, 2011 9:51pm Dec 31, 2011 9:51pm
Brokerages have their radars set for arbs, believe me. Now, the bad part is that you're going to get banned like card counting blackjack players do - for something, thats not strictly speaking illegal/forbidden. But I guess that makes you undesirable. Casinos, brokerages don't like to be taken advantage of, no matter what.
  • #172
  • Dec 31, 2011 11:55pm Dec 31, 2011 11:55pm
NQ, I think what this essentially boils down to is the Arb trades made by RR's system are not latency arb's in the way most of us have read about on the internet. Many latency arb strategies are very easy to detect because they either make a zillion trades that are nearly all profitable and close within a second. This is the "latency manipulation" that can be noticed by the broker.

I have seen the system in action and due to the time it holds many trades, and that several stop losses do get hit, distinguish this from the subsecond latency arb's so often mentioned in bucketshop's TOS. If you re holding a trade for minutes or even hours, how can the broker prove that you can see the other end of the interbank and are scamming them with this knowledge?

To me this seems very hard to prove in this instance.






Quoting nqm0t0
Brokerages have their radars set for arbs, believe me. Now, the bad part is that you're going to get banned like card counting blackjack players do - for something, thats not strictly speaking illegal/forbidden. But I guess that makes you undesirable. Casinos, brokerages don't like to be taken advantage of, no matter what.
Skype: heliosphan187
  • #173
  • Jan 1, 2012 6:43pm Jan 1, 2012 6:43pm
Quoting Dirtybrown
how can the broker prove that you can see the other end of the interbank and are scamming them with this knowledge?

To me this seems very hard to prove in this instance.
Its not difficult to detect, but they might only really feel it with larger lot sizes when they MIGHT choose to hedge the positions. They would persistently find it difficult to hedge the 'hot potato' they were passed (if they were a slow broker the price would have moved) In that way it appears more like an institutional broker position - and they have no problem spotting those. Even if they didn't actually hedge the trade they would see the price consistently moving away just after the position was opened - software can spot that easily.

If the trades are smaller, however - they might not bother, they certainly cannot accuse you of hyper scalping which is another frequently side effect of trying to 'front run' a slow broker feed

To disguise it you might be better waiting for the position to reverse - which it would in many cases - and then close at a loss on the 'slow' broker (and a win on the other broker) . Then you would have a mixture of wins/losses that would be more convincing. After all it MAY be only when they consistently lose that they take notice. This might also help in flattening the equity curve which will wander about as the net position moves between the brokers - its the equity 'pump' i mentioned some posts ago. In fact you might manage to stack the equity so that you LOST on the slow broker overall - then they would be happy

I don't think its necessary for them to 'prove' anything, unless they are looking to reverse trades or remove profits - an unscrupulous broker is more likley to start slipping you more to compensate - but why would they if you were one of the 95% of 'losers' out there
  • #174
  • Jan 1, 2012 7:00pm Jan 1, 2012 7:00pm
The other 'secret' to this is gonna be VOLUME - you gotta drive those commissions down. Look at the dukascopy commissions if you are running a larger account or larger volumes - they drop like a stone. Also take account of rebates in the calculations. I know the likes of GO did stop rebates, but i think they continued to pay to the announcement rebate end date so you should be ok if you keep a close eye on them
  • #175
  • Jan 1, 2012 7:15pm Jan 1, 2012 7:15pm
Quoting ianj1
Its not difficult to detect, but they might only really feel it with larger lot sizes when they MIGHT choose to hedge the positions. They would persistently find it difficult to hedge the 'hot potato' they were passed (if they were a slow broker the price would have moved) In that way it appears more like an institutional broker position - and they have no problem spotting those. Even if they didn't actually hedge the trade they would see the price consistently moving away just after the position was opened - software can spot that easily.

To disguise it you might be better waiting for the position to reverse - which it would in many cases - and then close at a loss on the 'slow' broker (and a win on the other broker) . Then you would have a mixture of wins/losses that would be more convincing. After all it MAY be only when they consistently lose that they take notice. This might also help in flattening the equity curve which will wander about as the net position moves between the brokers - its the equity 'pump' i mentioned some posts ago. In fact you might manage to stack the equity so that you LOST on the slow broker overall - then they would be happy
Very well said. That exactly is how you play smart and win in this arb game without drawing attention and having your profits cancelled on the slower broker.
  • #176
  • Jan 2, 2012 1:58am Jan 2, 2012 1:58am
so how many opportunities for arb does this system produce in say, a weeks time?

and other than the broker possibly getting upset, are there any actual legal implications to playing an arb strategy like this? thanks.
  • #177
  • Jan 2, 2012 7:22am Jan 2, 2012 7:22am
Quoting jowsom
and other than the broker possibly getting upset, are there any actual legal implications to playing an arb strategy like this? thanks.
I don't think, you people, understand what are you talking about..

Let's say you took sell on one broker and buy on the other when prices where

apart, then when they came closer together you close both trades,

imaging 2 charts one overlay on top of the other, and instruments have a

perfect correlation, So ones charts gets apart a bit you sell top one and buy

lower one (position same lot) On different brokers. When, eventually,

charts merge again

you close both positions, SO? you going to have possibly loss on 1 broker

and win on the other.. Next time you might have them flipped.. So?

you account not going to look like you think (All winners) NO. you going to

have winners and losers like any other typical account the only different is,

when you took stats from both accounts and apply them together only

then you going to see profits.. Now, how come individual broker can get

upset??? About what? Nothing unusual will be showing in statement.
Viewer Discretion Advised: Shall we shag now or should we shag later? :-)
  • #178
  • Edited 8:16am Jan 2, 2012 8:04am | Edited 8:16am
I have had a few PM's asking me what 'last look' is - so i thought i would explain.

This explanation directly refers to the institutional market, but MAY also apply to the retail market TO SOME DEGREE.

A pure ECN is when liquidity providers post bid/offer prices to the broker and the broker matches buy/sell orders against the bid/offers. The list of bid/offers is called the ORDER BOOK, and the 'best' bid (highest) and offer (lowest) is called the TOP OF BOOK - that is what you see in most retail price feeds - top of book prices (BTW: L2 is a more complete order book). In AUTOMATCHING the broker will match buy/sell orders against lying bid/offers (according to certain criteria including credit, price, time etc). When NO last look is active the broker will simply INFORM the liquidity provider (bank) that his bid/offer has been hit/matched and send the COMPLETED TRADE details. This process can be very quick (a few ms)

With LAST LOOK, just BEFORE the match - the buy/seller detail (price, size and sometimes account detail if it is not anonymous) will be sent to the liquidity provider to ASK them if they want to accept the deal. They are not really supposed to refuse many (very small %) of deals, but the broker WILL NOT PROCEED with the matching operation until the liquidity provider responds. An unscrupulous liquidity provide could dislike the deal but be reluctant to refuse it (they have accept quotas they have to meet) so they MIGHT choose to sit on the deal for a few hundred ms - to 'sour' the deal for the aggressive buy/seller hitting the banks price.

LAST LOOK is insidious, they don't need to refuse the deal - just delay it, to do damage. IF they eventually refuse the price- the buy/sell will go back into the matching pool to look at the next best bank price - imagine what happens if the same happens again - time to go on holiday !

None LAST LOOK brokers (or more exactly non last look PRICES) are quite rare, and almost always NOT the best prices (not top of book) so last look prices are very much a fact of life.

In the retail market i expect most prices are last look - as long as the broker sends the price to the liquidity provider who can veto the deal or delay the response

The ECN/STP question is moot - you can have ECN and LAST LOOK at the same time - and i am not sure if 'pure' ECN brokers with no last look actually exist in MT4 in any case

Another term that refers to the same process is AUTOMATCHING - means essentially means NO LAST LOOK

In the institutional world, none LAST LOOK brokers are EBS, Reuters and to a large degree HotspotFXi (maybe completely - i am not sure). There are a mix of LAST LOOK and AUTOMATCHING prices on Currenex/Lava(FxAll)

In Retail, its POSSIBLE that MB Trading and LMAX are true ECN, and may not last look - and LMAX claimed to me they did not last look - but beyond that i don't know. Dukascopy i SUSPECT has a mix of price types, and perhaps also Integral (FinFX and others).

The reason why it is mixed up with ECN terminology is because an ECN implies that normal mortals - you and me - can post bid/offer prices that appear on the order book. Invariable normal mortal prices will NOT have a last look (why should the broker care if normal prices are arb'd) so will likely execute quickly. Its the big liquidity providers (banks) that use it in most cases - and its their prices that are normally at top of book. Unless it is a big important trading platform (the 2 big boys are EBS & Reuters) a bank will be very reluctant to post good prices unless last look is offered by the broker. So no last look, no prices !
  • #179
  • Jan 2, 2012 2:03pm Jan 2, 2012 2:03pm
Quoting Ronald Raygun
Arbitrage requires buying in one market, and selling in another simultaneously. In this case, we are looking for two brokers with two different prices for the same pair. If the buying price at one broker is less than the selling price at another broker, then if you buy from the buying broker and sell at the selling broker, you turn a profit.
Hi,

Can somebody clarify this please.

Suppose i have following prices from different broker for EUR/USD buy - 1.2929 and EUR/USD sell - 1.3001 So i have an arb opportunity. I buy on one side and a sell on other. Sread is 2 pips. Price moves 5 pips up, i have +3 on the buy side and - 7 on the sell side. Where is the profit? Can somebody please explain it more widely, how is the profit generated.
  • #180
  • Jan 2, 2012 3:06pm Jan 2, 2012 3:06pm
i think it's really cool what ronald had done here, yet not profitable, then how jasonfielder get profit??when arb's what we talking about here.
i'm newbie @forex thing, just googling about this stuff and read ,want to ask ronald which want's better arb or corelation pairs ex eurusd buy ,another pairs sell (i don't know what pairs) that against eurusd and in the same broker and terminal,which want more profit?thank you.sorry my english bad
1 trader viewing now