The usual problem is of defining merit and testing it. The critical step that is almost always missing is testing the measure - matching it and results. Realit

The usual problem is of defining merit and testing it. The critical step that is almost always missing is testing the measure - matching it and results. Reality is much more messy than most everything used to decide merit - ultimately even third party judges can be biased or the sample taken might not be representative.