I'm sure that Hasbro play tests all of their toys. I suppose one potential problem with focus test groups is that it's always going to be a fairly limited sample of people that you're testing it on. While I'm sure it works most of the time, there may be times where the group tests different from how the overall market responds. It sometimes happens for other products like say movies.

e.g. the test audience for Apocalypse Now performed really poorly with the test audience who rated the film as "boring."

Or heck, how many times have you come up with what you thought was a really good lesson plan and then it worked really brilliantly with one class... but then totally bombed with your other classes (or vice versa)?