Right. But there's no comparison to, say, those pending free agents who took a long-term contract w/ their home teams, nor any real analysis of the problem: "Suppose player X will perform at Y-level if you resign him; if you let him go, he will perform at Y+1 for someone else. Do you match the other team's offer?" (i.e., is it better to overpay him to play for you, or to let him play for a competitor who will get full value out of him?) I guess this is why the GM's office is full of people scratching their heads hoping they guess right.