Well, yes - I'm just talking about the cost of a wrong hire. Sure, if you have a bad apple, get rid of it. But it's better to ignore a few good apples if that means you reduce the likelihood you get a bad one.
Talking with friends at Google, the perception there is that the cost of a false negative is larger than the cost of a false positive. The catch is that you need to have an agile hiring process, AND actually let go of people who don't carry their weight. Most companies fail at both points especially the second, until hard times come and investors demand rolling heads.
