There will always be problems with admissions and selection processes. There are many people who do LOTS of practice for UMAT and barely get in, and a lot of people who hardly do any practice at all, and barely miss out. If UMAT was purely a measure of 'common sense' then clearly, the latter group would be more deserving. Of course, there's no way to stop this. Similarly, ATAR is not good measure of academic merit either, as you guys have said, at least at a very high level. Anyone who gets above 95 or so is prooobably got enough brains to handle med, but that doesn't mean someone who didn't get above 95 doesn't--again another flaw. And yes, everyone has their own strengths, so if you get high scores in areas like maths and physics, but you're really bad at rote learning, you may not be suited for medicine even if you get in. Such things can't really be solved with out current system (if ever), so I think many people's frustrations are justifiable to an extent.
Personally, I think equal weighting given to atar, umat and interview is about as fair as we can have it (for reasons I outlined above, that every criteria is flawed in its ability to measure certain things about the candidate). Some may argue that interview (communication skills etc) are important and may be given more weighting, but communication skills can be learnt and improved, and not only that, it takes by far the least amount of time to complete. Because of this, I think based on effort alone, it would be unfair to give interviews a larger weighting as well.