It is a demonstrable point in the UX field that initial negative feedback is not as helpful as you might think. In the short term, humans typically react negatively to change they don't themselves initiate.
The real question will be how folks feel about it in a few weeks or so, after the shock of unannounced change wears off.
You have a valid point.
However, there are other considerations.
Initial negative feedback WITHOUT data points is just a gut reaction, much as you say. "I don't know why I don't like this, however RAGE!"
However, an initial reaction , such as not liking blue text on a bluer background, missing functions, etc (just general examples) is valid initially and over time. Its actual feedback.
Then of course there are ways of implementing change to reduce the "shock". Unfreeze, Change, Refreeze. Probably not a process you want to use in a widely vocal internet arena, you would be overwhelmed by getting folks to buy into the concepts, but some of the concepts could be used. Post a thread in Meta "Going to Test Layout?"
Finally, it depends on goals, if the change initiator wants the change, for reasons (speed, cost savings, preference, etc.) then the longer you go past the initial change, the easier it is to keep it. As it becomes the new norm, people get tired of mentioning it, or new folks don't even know there was a change.
Anyway, not a show stopper, thanks for the discussion with us [MENTION=177]Umbran[/MENTION].