Wii should have tested!
Nintendo is willing to replace some 3.2 million Wii Remote wrist straps at an estimated cost of $1 million because the straps have widely been reported to break during regular gameplay, in some cases causing the remotes to crash into TV sets and even injure other players. There is a web site called Wii Have a Problem that is dedicated cataloging Wii-related accidents.
Personally, we had a Wii night last night at my place and someone-who-shall-not-be-named did hit the ceiling with his Wiimote (we discovered that my living room isn't actually high enough to jump to smash a ball in tennis) but, thankfully, the said person didn't suffer this guy's fate :)
I was recently talking about how important it is to carry out user testing during product development at my Memo to the CEO talk at Flash on the Beach. This is a perfect example of a problem that would have been discovered earlier had Nintendo carried out contextual user testing with its European and American users (I'm not sure how big a problem the wrist strap issue is in Japan). Contextual, meaning in their living rooms (ironically, this exact scenario is depicted in the North American Nintendo ads: "Wii would like to play!"). So basically, what I'm saying is that Nintendo didn't play with their Wii enough. (Oh, no, it's doesn't get old does it?) :P
On a serious note, this is an example of how contextual user testing could have saved Nintendo a million dollars. How much will user testing save on your product?