LOL. The fact that some dates are falsified through contamination does nothing to undermine the method. It's like saying that your laptop never works because it doesn't work when you run it right next to the electromagnet they use for lifting cars. When a method which, in theory, should be very robust starts giving completely nonsensical results, the first thing you do is not jettison the method completely, the first thing you do is check to see if there's any reason why the result would be that way. For example, the reservoir effect makes carbon-dating marine animals a basically futile task. This doesn't mean, however, that it doesn't provide robust results when examining land plants. Indeed, in many of these cases, we'd need to be wrong about the rate of radioactive decay in order for the method not to work. Obviously, it's far more likely that the samples have been somehow corrupted, and that's something that needs to be explored. Now, if no source of such corruption can be found, then we have a problem. But to date, the answer has never been "our understanding of how radiometric decay works is totally wrong", but rather "Oh, we're getting a wrong result because the sample has been contaminated".