Triangular Arbitrage

I've been reading about triangular arbing and was surprised to find that the cross of two majors doesn't always equal what the calculation would suggest it should. I would have thought that brokers would offer the cross as a direct calculation of the two so that arbitrage opportunities wouldn't exist. From what I have read at least at some point in history this wasn't the case.

This immediately gave me the idea of buying gbpusd and selling eurusd, so taking a short position in eurgbp but hedging that trade with a long in eurgbp itself. Should the rate offered for the cross go out of skew with the calculation of eurusd + gbpusd then there would be a risk free profit opportunity.

Does anyone have any experience of trading this way, is this still possible and is my idea feasible? Thanks.

Yes, I can see your reasoning, its very trendery. Its like lots going on around you but you always have few bucks in your pocket.

Cant remember the thread names hugh,but you could search them out, were some traders are doing somthing similar with some sucsess or they were then,sorry cant be more helpfull

Algos absolutely crush any difference seen.

Brokers and banks also often markup the cross currency spreads just enough to buff out any advantage as well.

The firms still doing it rely on different ECN's, automated execution, and low latency trading. They'll sit on one ECN for the common currency and monitor the other for the cross currency difference.. otherwise the banks rule this market. Rightfully so though, as the banks are making the market in most cases and someone arb'ing some free profit from their quotes comes out of their pockets.

Thanks for the input everyone. I noticed once after the nfp announcement that there was a difference of about 30 or 40 pips between Alpari UK and VantageFX UK. I guess news spikes may offer some good opportunities but on an ongoing day to day tick to tick basis it would be generally quite difficult to capture any significant gains. Most of that difference was probably due to spread widening anyway.

Hugh,

The comments so far are on target with my understanding. Generally speaking arbitrage (true arbitrage, not picking off slow brokers) is a game of the fastest and is typically out of reach for retail traders.

But those points aside, many/most real edge based trading strategies are situated on top of the arb concept (taking advantage of real inefficiency). If you're interested in learning more about the topic of Triangular Arbitrage including the mechanics of forming synthetic pairs, some of the realities for retail traders hoping to cash in, and a possible way to determine which pair is out of balance (which may spur on some system development ideas) take a look at the link above. Also, let me suggest a few threads you might find interesting.

You'll note that there were / are significant technical hurdles to overcome in implementing an arb platform from reading the first link below. I'm not sure if the OP has overcome them yet or given up.

http://www.forexfactory.com/showthread.php?t=331106

http://www.forexfactory.com/showthread.php?t=310349

http://www.forexfactory.com/showthread.php?t=223946

http://www.forexfactory.com/showthread.php?t=64423

(This last thread - note how the OP requested the thread to be deleted! Always a good sign )

Excellent, thanks for your help.

What about arbitrage opportunities when the cross reacts to a big round number or other technical? Surely that is something that's exploitable?