What I see reading that is that the OpenRC was not seriously considered. There are a bunch of claimed requirements that appear to rule out OpenRC, but I don't see those requirements tracked back to any benefits. Perhaps the justification for those requirements is obvious to those who made the decision, but it isn't obvious to me.
Taking the requirements in turn:
* Lack of integration with kernel-level events to properly order startup.
So what? OpenRC has dependency built in and the added improvement of integration with kernel-level events would bring only a very minor improvement.
* No mechanism for process monitoring and restarting beyond inittab.
In my experience, this is solving a non-problem. I don't experience processes dying and needing an immediate re-start without any other action.
* Heavy reliance on shell scripting rather than declarative syntax.
* A fork and exit with PID file model for daemon startup.
Not sure what advantage this brings.