I'm not sure that was bunk in Nokia's case. When they realized that Symbian had to go, they also realized that the vast bulk of the company was organized around that one OS. And if sources are to be believed, most of that deadweight was fiefdoms jealously guarded by entrenched groups, which meant they were not going to be conducive to "pivot" to a new direction. Now they had to shed that humongous deadweight
and produce a competitive "modern" smartphone. I think it's entirely possible that in that situation whatever resources you have left, you may well have no choice but to bet the farm.
Nokia may have used Android to force Microsoft into a buyout, but I doubt they were going to seriously pursue it, for one big reason: that would have entailed compulsorily licensing Google Maps, which is a direct competitor to their own mapping division. You may think that doesn't sound so bad, but consider this:
1) It dilutes the billions they paid for Navteq.
2) As the SkyHook lawsuits have shown us(IIRC) Google demands all location data produced by Android devices to improve their own geo services. As such, Nokia would have ended up building devices that end up improving a competing service.
3) Considering the mapping division is one of only three divisions Nokia held on to, it's undoubtedly important to them long term. I am not sure, in their eyes, the pros of having an Android phone out there would have outweighed the cons of improving a competing mapping service.