I do not think you fully understand how arbitrage works. The make it or break it factors are transaction fees & slippage. There are many ideal arb oppotunities involving index sector options & ETFs. However, when you really break it down and consider the real world of trading, you will see that it's a net loser most of the time. The risk is in slippage. You are looking at very low volume ETFs. What you do not realize is that they are being arbed, but the price innefficiency has to be very significant before a hedgie would step in. They have already calculated for the slippage. Not to say that you cannot try it. You may have good results with small shares. But the scalability of it will be very limited. In other words, not really worth it. Also, if you were to practice REAL arbitrage regarding ETFs or sector index options, you need to educate yourself on how to actually convert the ETFs into shares. That involves more additional costs, administrative & execution.
i dont quite understand this. executable? you mean that when i see a change on IB i can execute on it? i am under the impression that IB sends a max of 1 change per 50milliseconds, say eSignal or something sends a market change to me 50microseconds after it happens. in both cases (hypothetically) it takes 12milliseconds for the packet to reach my server. so in the eSignal case i would be act on a arb 49950microseconds earlier. if the above situation is the case. am i wrong? if anyone at IB who knows the limitations of the data distribution system could answer. you might save me money from the data services.
well i was exaggerating for sake of example to show a better data processing / transmission time would improve arbing. i know the kernel quantum on most modern computers is about >=50micros. as long as the response time from when the data provider gets the change from the exchange to the time they get around to transmitting to me is < 50000 microseconds, it should outperform IB's data? no?
Another thing you need to overcome are SEC Fees. Right now they are low at .0000153 but when they go back up to around .00005, there goes your small profit. You can do the math yourself to figure out how much they will cost you on the SSO and SPY arb.
No - IB's quote is the price you are executed at. If that is out of sync with your other data providers, that's a problem of an even greater magnitude.
Meaning it's of little consequence if your quote server can identify arbs but IB can't execute. Anything "faster" than your broker's feed is moot.
you think a hypothetical ~50000microsecond advantage is moot when it comes to arbing? seeing an arb opportunity on an exchange earlier means sending your order to the exchange earlier. when liquidity is first come first serve, its quite meaningful.