Hi @MGL_Data2
Thank you for taking the time to share your observations.
I would like to clarify a few points that may help everyone better understand the execution flow.
The scenarios you’ve described are not caused by the Algo strategy itself, but are a consequence of the latest exchange regulations, under which API-based orders are required to be placed only as LIMIT orders. This means that order execution is now dependent on market liquidity and whether the LIMIT order gets filled.
For multi-leg strategies such as Credit Spreads, the long (hedge) leg needs to be executed first so that the broker RMS can unlock the hedge benefit. Only after this can the short leg be executed with the reduced margin requirement.
If the first leg remains pending and the second leg is placed, the broker RMS may reject the second order due to insufficient funds, since the hedge benefit has not yet been released. A similar flow applies during exits, where the short leg needs to be exited successfully before the long leg can be closed. If the first exit order remains pending, the subsequent order may again be rejected as per the broker’s RMS rules stating that it’s a part of the hedge strategy.
To improve the overall execution experience within these regulatory constraints, we have introduced the Retry Mechanism as the first step. Rather than immediately proceeding, the Algo now retries (for exit) pending orders before moving ahead, significantly reducing such scenarios compared to the earlier implementation. To learn more, refer to the post here: Retry Mechanism: Stratzy’s Most Requested Feature Update
That said, we completely understand the expectation that user intervention should be minimal. Improving execution reliability continues to be one of our highest priorities, and we will keep enhancing the order management system while remaining compliant with the latest exchange regulations.
If you notice any difference in P&L, we request you to please reach out to the support team of the platform through which the Algo has been deployed, along with the relevant details. We will be happy to review the specific instance and provide further clarification.
Lastly, the orders displayed under History represent only the orders placed by the Algo. We recommend avoiding manual intervention during the execution process, as this can lead to mismatches between manually placed orders and Algo-managed positions. However, the PnL History of the Algo gets updated even though you intervene manually.
I appreciate your feedback and patience as we continue improving the platform.