I'm going to be running an automated strategy on ZG and ZI using stops to enter. Given some filters on the system there's a possibility of sending a stop order where the stop price has already been penetrated. Since I know from past experience ECBOT will reject a stop where the price has already been hit I am considering either: - send a stop limit (which is simulated by IB) where the limit price is a small amount above the last price - change my code to detect this and use a market order My preference would be to use stop limits since changing my code could cause unintended problems and would require re-testing the strategy. Changing to use a stop limit would be a small change to the order entry logic and much easier to test. Does anyone know what the ECBOT limit is on how far away a limit order price can be from the current price? I don't want my stop limit to reject or cause one of their $2000 charges. Can someone verify that when IB simulates a stop limit it waits for the stop price to hit then submits a limit order. Seems like it would be a no-brainer (how else could they simulate it???) but I don't want to find out at a cost of $1000 that I'm wrong....
you also have another choice now ... "comex" on globex for your orders ... ( GC + SI ) a stop limit order looks like the right approach ... only way that hurts you is if the metals gap hard and do not come back to your stop limit price
Comex seems to have wider spreads on GLOBEX when I've looked in the past. Thanks for the idea though. I'm still concerned about the limit order somehow not being inside the "correct" limit price band, I've seen things about a $2000 charge you get hit with for doing that.... I'd hope the metal wouldn't gap much in the 5 seconds or less it takes to process the strategy and send the order....but with my luck it will happen....