You can deploy and go to sleep if you have monitoring and alerting and someone getting paged. It shouldn't be a human monitoring for issues anyway, so the only reason to choose 9am over bedtime should be that you don't want to risk a late night page, not that someone will actually be checking up actively on the deployment.
For a big launch, you can have your entire team looking at issues in realtime, rather than a small number of people responding to pages in the middle of the night. I think if it's worth it to you there's a big difference.
That is, if you know what to monitor. When you launch an entirely new feature, you often don't know all the ways it will break.
And it's not trivial to set up a correct alert for this one. Simple HTTP 5xx threshold wouldn't work because it's high enough to not wake you up whenever cloud provider restarts Postgres, it's too high to catch this. You need either per-endpoint failure rate alerts or something else more clever.
