Going into the close this afternoon, I was short a May 930/960 call spread in AMZN. When I logged in near the close to close the position, I discovered that the position had already been closed for $30.55 (likely at the natural?). I'm not fussed about the auto-liquidation itself as my long leg was at risk of being pinned, but surely this could have been closed for $30.00 or even $30.01. It's only $0.55 but do I have a valid beef?
Having seen my share of auto-liquidations, I have observed as much on the other side as at the mid. FWIW and all that. (Oooop: IB/TWS...)
Any time you are dealing with IB, you should factor it into your equation, just call it the IB spread, assume you will get the worst price possible if the shares even show up at all, and there is a high likelihood you wont even be able to log into the software when you need to exit the position. Backtesting systems based on TWS you should just automatically assume you will take a 20% hit on both sides, so any system that doesnt produce more than 40% per trade on average, you should just assume TWS will find a way to fuck you.
Going into the close this afternoon, I was short a May 930/960 call spread in AMZN. When I logged in near the close to close the position, I discovered that the position had already been closed for $30.55 (likely at the natural?). I'm not fussed about the auto-liquidation itself as my long leg was at risk of being pinned, but surely this could have been closed for $30.00 or even $30.01. They closed it at the "natural" do I have a beef ? No
IB does not close positions precisely at the close. They allow themselves a certain time window to do this. http://ibkb.interactivebrokers.com/node/199 You would need to look at what happened to the bid and ask prices during this window.
I was filled 5 minutes prior to close. From a customer service perspective, it would have been reasonable to start with an offer of $30.00 and penny up vs simply a market order on the spread. I wonder what their threshold is for doing this from a pin perspective as I've let many spreads simply expire with each leg getting assigned/exercised. Oh well--lesson learned. Going forward I may simply avoid the hassle and just exit the position earlier.
If I remember correctly, IB's auto-liquidation process is automated. The program is not going to bother penny up orders to help you out. It's just going to put in a market order to flatten the position, which basically fucks you in non-liquid markets. I still remember the horror stories of IB liquidation during the flash crash. I have never used IB for that reason alone.
I understand using a market order for products with undefined risk. But one would have hoped their algorithms would be sophisticated enough to understand that the max value of the spread is $30.00. I guess that would be asking too much of them.