|
@patrick.lapointe We currently have two separate OCA related topics. Let's keep your issue in this topic you created. Which of the three are you using? It sounds like you may have a possible "overfill" situation. Two orders want to sell one unit each so that your long position of one unit would become a short position of one unit Also, is this just an OCA with three or four independent orders, or is this a bracket order with one or more parent/child relationships? The screen grab in your reply /g/twsapi/message/54115 suggests that you have two orders with exactly the same trigger price in OCA:
Not quite sure what you are trying to do, but I have the feeling that is part of the problem. ´³¨¹°ù²µ±ð²Ô ? ? ? On Tue, Apr 22, 2025 at 12:20 PM, <patrick.lapointe@...> wrote:
|
|
Hi,
?
Pretty sure both messages are the same issue.
?
The STP LMT 2.24$ order is active right away.
The LMT 2.24$ order becomes active in 20min(Now+20min).
?
To explain why several OCA orders, I'm gathering stats on different OCA group scenarios.
?
OCA group is supposed to cancel all other orders if 1 order is filled right?
So how is it possible to sell more than 1 order within 1 OCA group?
?
?
Thanks |
|
You might want to read up on OCA details. Exchanges generally do not have native OCA implementations, so IBKR simulates them within their system. Only OCA types 1 and 2 ("with block") assure that no overfill can take place by pacing one and only one order at the exchange at a time. But possibly at the expense of reaction time when one of the other orders in the OCA triggers. OCA type 3 routes multiple orders to the exchange and might result in overfill. ´³¨¹°ù²µ±ð²Ô ? On Tue, Apr 22, 2025 at 02:45 PM, <patrick.lapointe@...> wrote:
|
|
You control the OCA behavior with the field in each order object. The default value is 0, which let's IBKR pick the mode. Check the Order objects you receive from openOrder() callbacks or set the field explicitly to:
1 for "Cancel all remaining orders with block."
or
2 for "Remaining orders are proportionately reduced in size with block."
?
Type 1 sounds like what you want to go with.
?
´³¨¹°ù²µ±ð²Ô
?
On Tue, Apr 22, 2025 at 03:10 PM, <patrick.lapointe@...> wrote:
|
|
The number of orders in the OCA does not matter (other than the max number of open orders for one instrument which was 20 at one time). I am still not sure what the IBKR OCA implementation does when it sees two different orders (LMT and STP LMT) with the same trigger price. Even if one of them has a time condition. ´³¨¹°ù²µ±ð²Ô ? On Tue, Apr 22, 2025 at 03:24 PM, <patrick.lapointe@...> wrote:
|
