Alright, here's a question for the mathematical geniuses out there. I guess it may not take a genius. Just an argument with a buddy.
If you are doing squares for March Madness (or any sports squares for that matter) do you have worse odds for same numbers (ex. 3 for winner, 3 for loser) if you take ONLY the final score of game? Meaning if a game goes into OT you take that score and not the final tying score of regulation.
I was stuck with 3 and 3 so I feel I have worse odds of winning with our group's scores coming from the final score including OT. It may be marginal but I feel I have an argument here.
Without having all the data to support it, assuming you are only considering the final score (quarter or half time score payouts will be different); I believe the answer varies depending on the which teams are playing and how big a talent/performance difference there is between them. For example, if the two teams are equal on a neutral court (NCAA tourney), you could expect some sort of normal distribution of final score outcomes for Team A around 0, but not 0. So having 1 point difference is most likely (win or lose by 1), then 2, then 3, etc. You only care about the absolute value in this case. So your answer is having the same number is bad, as the sum of the probabilities of wins by 10, 20, 30, etc. will be a lot less than 1, 11, 21, etc. - especially since 10 and 11's probabilities are about the same. But where 0 is an option at half, that becomes a slight favorite.
But then there is a scenario where one team is 10 points better than another. Your normal distribution curve moves + or - 10 spots and having the same number becomes the best probability for the end of the game, but probably not half time.
The best thing to do is look at the Vegas odds to see what the expectation of the match up is. For example if your squares were on the Florida (-9.5) vs ETSU game, you should be happy. If however, it is on Maryland (-2) vs Xavier, you have low odds of winning.