Almost all of the orders I make that are marketable are reported as filled via "smart". I assume, but perhaps I should not, that these orders are filled by IB's market maker branch, Timber Hill and they are able to do this profitably. This is not a bad thing. TH provides liquidity at the NBBO or better. I may not always be happy with the fills I get, but after reviewing the price action, I have never found a fill that looks to be incorrect. They are always fair and that is all that I can ask for. The fills I get on orders reported as filled via "smart" are almost always for sub penny amounts. So, are you saying that IB's restriction on routing via the API is not because smart orders are more profitable for IB? They should be much more profitable. If so, then why does it cost more to direct an API order?
Today, I made 1,170 trades = 183,461 shares: NYSE 81.7% ARCA 4.1% ISLAND 3.0% BATS 4.6% NSX 0.6% DIRECT 1.5% SMART 4.6% Over the past year SMART has ranged from 3.0% to 30.0%... But NYSE has clearly been re-gaining market share all year 2010... Though different stocks may have a different profile. What many inexperienced traders seem to miss: If you use a Limit Order any amount away from the NBBO... You are forcing an ALIGNMENT OF INTERESTS... Between you and your broker... You BOTH benefit if your order is executed at the Limit Price... And since there is very little price elasticity... IB doesn't much care where it's executed... Because IB only gets paid IF it's executed... the price is hard-coded... So over my 1,170 fills probably > 90% were the fill I wanted. If you mess around with any "marketable" order... YOU... YOU THE TRADER... YES YOU... Are creating a Conflict of Interest... Between you, the broker, and the exchange... And you deserve whatever you get in spades... So bend over and take it like a man. As for "API orders must route SMART"... It's really "unbundled" orders must route SMART... Because I cannot imagine an API user using "bundled" pricing... And as explained above... Routing is not a big issue for Limit Orders... They almost always end up at the Primary Exchange... And if you wanna play Algo and Rebate game... Go trade elsewhere.
I think that if you are willing to pay the spread, even if it is just a penny, the order is not likely to ever hit a public exchange unless you are using routed orders. I use a lot of stop orders for exits and I am sure that I am leaving money on the table. But, I haven't been able to find anything that backtests better, so it is ok for Timber Hill to provide the liquidity. The odd thing is that I haven't been able to find benefit in peg orders. From the description, I expected them to get filled faster than a limit order, but is not the case in practice.
This is likely a question for the IB guys here, but maybe others know from experience. Can a Peg-to-Market order, available on orders directed to Island, which generally Buys on the Offer minus an offset and Sells on the Bid plus an offset be entered without an offset? Support materials are contradictory on this: Users' Guide indicates "No", while Knowledge Base video suggest "Yes". If it can be entered without an offset, does it rollback when the offer drops (on a Buy order) or when the bid rises (on a Sell order)? Last, can it be entered with a "cap", so that one can set a limit on the maximum/minimum price they're willing to accept on a buy/sell order? If so, how? The support materials on this issue for this order are entirely unclear. Thanks in advance.
Maybe IB provide superior executions overall, but if we talk about pegged orders in particular, IB's pegged orders that dont roll back definitely give your clients "equal or worse" executions than exchange pegged orders, always, every time. A similar study on execution quality of pegged orders would be embarrassing for IB. I think IB only offer these pegged orders that dont rollback to increase number of executions, and therefore commission dollars. Think about it, if the pegged order doesn't roll back and keeps getting more aggressive, its much more likely that the order will get executed (at a worse price for the client of course). More executions equals more commission dollars for IB. Its that simple folks.
Another uninformed post from someone thinking the use of these pegged orders are extensive in the system and the way they trade is the only way to trade. They were specifically designed for large institutions to the specifications they requested.
dear Def, can you confirm/answer my question above? can I set zero offset for REL order while allow a postive discretionary amount? thank you for your help on this forum. it is great to have someone inside IB. as someone who pays IB 5K or more of commission each month, i have to say my experience overall is very good with IB. njrookie
If they were designed to accommodate the needs of your clients generally, they would be used extensively. And while they wouldn't be the only way to trade, they would certainly be a very sensible way to trade, which unfortunately is not to the case given the way they're currently configured (i.e., without rollbacks). I note with some interest that you and other IB folks here keep falling back on the "this is the way the large institutions wanted them" explanation for the fact that these orders do not serve the majority of your clients' interests and, instead, serve IB's, which are, of course, to generate maximum revenues. Since you keep relying on this explanation with absolutely no support, would you be good enough to explain why any institution would want to pay more on its buy orders and receive less on its sell orders than it would pay and receive were these orders designed with rollbacks? Because this is precisely the overall result of the way these orders are currently designed, supposedly "to the specifications [your large institutional clients] requested". Stated simply, you're asserting that large institutions designed orders that cost them more and benefit them less than the same orders designed with rollbacks. I trust you'll understand that this assertion is not, at least on its face, terribly credible. I look forward to hearing from you on this and hope that you'll push for a rational redesign of these orders such that they incorporate at least the option of rollbacks, whatever the rationale for their original design may have been. Thank you.
1. Yea, you're right, there seems to be a lot of us "uninformed posters" on this thread and on previous threads dating back to 2005, all wanting pegged orders that roll back. We should all just "get a clue" i guess... 2. If pegged orders aren't "extensive in the system", how come they were "specifically designed for large institutions"? Contradiction? Plus, who cares how extensive they are? Are you saying that "IB doesn't care about this small group of clients who use pegged orders because its a small group of clients"? Not good. 3. Noone's buying the large institutions story anymore. Why would large institutions want an order that would harm them. Its not logical. Even if it were true and those large institutions were stupid enough to want that years ago, its no reason for the rest of your clients to suffer for years to this day. You could always provide rolling-back exchange pegged orders AS WELL AS your non-rolling-back simulated orders, and give people a choice... but you don't... and you don't explain why you don't either. Doesn't look like this issue is going away this time, like it did in previous years. And you only make matters worse for IB with these "more-of-the-same", evasive comments. Maybe you should just do the right thing and give these folks exchange pegged orders that roll back like every other direct access brokerage out there, and remove your crippling, outdated, simulated pegged orders entirely so none of your "uninformed" clients accidentally shoot themselves in the foot using them. That would be better than trying to dance around the issue with one-liners and invalid excuses, and hoping it goes away for another few years.