Support Board
Date/Time: Mon, 25 Nov 2024 12:21:25 +0000
Post From: Is scale in / scale out client or server managed?
| [2024-03-02 16:15:25] |
| User601931 - Posts: 11 |
|
Do not think of it as the "Scale In/Scale Out" is stored on the server. What is stored on the server is the Target and Stop. When you Scale In/Scale Out the quantity of those Targets and Stops (on the server) are adjusted based on your Scale In/Scale Out settings. Got that. But is the client sending the instruction to the server to adjust the Q, or it is server managed? Because if the client is dead, it can't tell the server to change the Target/Stop quantities. If you have resting exit orders to scale out, then there is a risk of overfill. See my example below. We are not understanding the above statement. Are you saying that you have seen a specific issue? Yes, always in sierra charts. Please try the following: Assume the market is trading at 4000, and 1 tick = 0001. 1. Submit a market buy, Q = 5, with attached orders. The target is 20 ticks offset and the stop is 20 ticks offset. 2. In the order book & position. You should have: -> a long position of Q = 5 at price = 4000. -> one sell limit order at 4020, Q =5. Call this "Order A" -> one sell stop order at 3080, Q = 5. Call this "Order B" 3. Assume the market moves up slightly and you want to exit 1 lot at 4010 (ie: scale out) 4. Place a sell order for: Sell Q = 1, Limit, Price = 4010. Call this "Order C" 5. In the order book & position, Sierra charts will show: -> position: long, Q = 5, price = 4000. -> Order A: one sell limit order at 4020, Q =5 -> Order B: one sell stop order at 3080, Q = 5 -> Order C: one sell limit order at 4010, Q = 1 6. Sum up of the sell limit order quantities = 6. Therefore, if I loose the network/power and my local SC instance dies, I have 2 exit limit orders sitting on the server, with a total Q = 6, which is greater than my position quantity of 5. From my observation, Sierra Charts only adjusts the target/stops quantity on change in POSITION, not when you submit a new exit order (ie: order to scale out). So a disconnect can mean overfill rather than flatten because the server Q never got adjusted. Unless the adjustment is server managed, or I am missing a setting to force a recalculation of the resting target/stop attached orders based on sum of all exit order quantities (by ordertype). Thanks. |