Everyone says the thing they're working on is critically important. Who's right?
More work gets done for less if you wait until the 11th hour and fix the real problems last minute rather than fix everything ahead of time, much of which will turn out to not have needed fixing.
Yeah there's risks involved but at the limit it makes some sense.
Who is right is the wrong question here (not that your point is wrong - it is correct in some situations but not the one I'm talking about). This is a case where the features we need for the release were planned in advance and management signed off on them - by definition getting the feature done is right (even if it turns out customers don't want it, at this point we have committed as a company). However there are always a few bugs that become last minute stop ship issues that should have been prevented long ago.