The Definition of a Bad Poll Question

This question was asked by ESPN.com about the Ohio State matchup against Michigan.

11) Which of these role players will have the greatest impact?

32.6% WR Adrian Arrington, Michigan

22.4% RB Chris Wells, Ohio State

19.3% WR Brian Robiskie, Ohio State

17.5% K Garrett Rivas, Michigan

8.2% K Aaron Pettrey, Ohio State

This question doesn't mean a whole hell of a lot because Ohio State has three options and Michigan has only two options. Assuming that most people choosing are partial one way or another for a team, this question had to go to a Michigan player because their vote was only split two ways and OSU's was split three. If you break it down by team, it looks a bit different. All of a sudden the question is split almost 50/50 with the slight edge going to Michigan.

Regardless, I think it is safe to say that if there were three Michigan choices and only two OSU choices that an OSU player would win. The conclusion is that the results of this question are stupid. How about someone take a statistics course before they set this up next time?