Document regarding Postgres and replication methods describe “Both CDC and xmin are the most reliable methods of updating your data” can you expand on why Standard should be avoided?

Based on the knowledge sources provided, I can explain why CDC and xmin are considered more reliable methods for updating data compared to the standard replication method in Airbyte for Postgres sources. The standard replication method, which uses a user-defined cursor, has some limitations that make it less reliable: 1. Inability to capture deletions: Standard incremental replication cannot correctly transmit deletions from the source to the destination. This is because queries executed during syncs cannot return deleted records [Explore Airbyte's Change Data Capture (CDC) synchronization](https://airbyte.com/tutorials/incremental-change-data-capture-cdc-replication). 2. Missed intermediate changes: Standard replication only captures the state of the source database at the moment the sync is executed. Any intermediate changes, such as multiple updates to a single record between syncs, are not correctly captured [Explore Airbyte's Change Data Capture (CDC) synchronization](https://airbyte.com/tutorials/incremental-change-data-capture-cdc-replication). 3. Requirement for a suitable cursor field: Standard incremental replication requires the source data to have a suitable cursor field (e.g., updated_at) to keep track of which records have already been replicated [Explore Airbyte's Change Data Capture (CDC) synchronization](https://airbyte.com/tutorials/incremental-change-data-capture-cdc-replication). In contrast, CDC and xmin methods offer more reliable data updating: 1. CDC (Change Data Capture): - Captures all changes, including deletions, by reading from the transaction log [Postgres Replication Methods](https://docs.airbyte.com/integrations/sources/postgres#postgres-replication-methods). - Doesn't require a cursor field [Explore Airbyte's Change Data Capture (CDC) synchronization](https://airbyte.com/tutorials/incremental-change-data-capture-cdc-replication). - Recommended for very large databases (500 GB or more) [Postgres Replication Methods](https://docs.airbyte.com/integrations/sources/postgres#postgres-replication-methods). 2. Xmin: - Uses the xmin system column to track inserts and updates [Postgres Replication Methods](https://docs.airbyte.com/integrations/sources/postgres#postgres-replication-methods). - Doesn't require a well-defined cursor candidate [Postgres Replication Methods](https://docs.airbyte.com/integrations/sources/postgres#postgres-replication-methods). - Suitable for replacing full-refresh syncs [Postgres Replication Methods](https://docs.airbyte.com/integrations/sources/postgres#postgres-replication-methods). However, it's worth noting that xmin also has some limitations, such as not supporting row deletions and potential issues with transaction ID wraparound [Troubleshooting Postgres Sources](https://docs.airbyte.com/integrations/sources/postgres/postgres-troubleshooting#troubleshooting-postgres-sources).

Based on the knowledge sources provided, I can explain why CDC and xmin are considered more reliable methods for updating data compared to the standard replication method in Airbyte for Postgres sources. The standard replication method, which uses a user-defined cursor, has some limitations that make it less reliable: 1. Inability to capture deletions: Standard incremental replication cannot correctly transmit deletions from the source to the destination. This is because queries executed during syncs cannot return deleted records [Explore Airbyte's Change Data Capture (CDC) synchronization](https://airbyte.com/tutorials/incremental-change-data-capture-cdc-replication). 2. Missed intermediate changes: Standard replication only captures the state of the source database at the moment the sync is executed. Any intermediate changes, such as multiple updates to a single record between syncs, are not correctly captured [Explore Airbyte's Change Data Capture (CDC) synchronization](https://airbyte.com/tutorials/incremental-change-data-capture-cdc-replication). 3. Requirement for a suitable cursor field: Standard incremental replication requires the source data to have a suitable cursor field (e.g., updated_at) to keep track of which records have already been replicated [Explore Airbyte's Change Data Capture (CDC) synchronization](https://airbyte.com/tutorials/incremental-change-data-capture-cdc-replication). In contrast, CDC and xmin methods offer more reliable data updating: 1. CDC (Change Data Capture): - Captures all changes, including deletions, by reading from the transaction log [Postgres Replication Methods](https://docs.airbyte.com/integrations/sources/postgres#postgres-replication-methods). - Doesn't require a cursor field [Explore Airbyte's Change Data Capture (CDC) synchronization](https://airbyte.com/tutorials/incremental-change-data-capture-cdc-replication). - Recommended for very large databases (500 GB or more) [Postgres Replication Methods](https://docs.airbyte.com/integrations/sources/postgres#postgres-replication-methods). 2. Xmin: - Uses the xmin system column to track inserts and updates [Postgres Replication Methods](https://docs.airbyte.com/integrations/sources/postgres#postgres-replication-methods). - Doesn't require a well-defined cursor candidate [Postgres Replication Methods](https://docs.airbyte.com/integrations/sources/postgres#postgres-replication-methods). - Suitable for replacing full-refresh syncs [Postgres Replication Methods](https://docs.airbyte.com/integrations/sources/postgres#postgres-replication-methods). However, it's worth noting that xmin also has some limitations, such as not supporting row deletions and potential issues with transaction ID wraparound [Troubleshooting Postgres Sources](https://docs.airbyte.com/integrations/sources/postgres/postgres-troubleshooting#troubleshooting-postgres-sources). In summary, while standard replication is not explicitly advised against, CDC and xmin are considered more reliable due to their ability to capture a wider range of data changes and their reduced dependence on specific database structures like cursor fields.