Super Bowl Scores Predict Presidential Elections
Consider these scores for recent Super Bowls for professional football:
1992: Washington 37 Buffalo 24....Clinton (D) was elected President
1996: Dallas 27 Pittsburgh 17....Clinton (D) was elected President
2000: St. Louis 23 Tennessee 16....Bush (R) was elected President
2004: New England 32 Carolina 29....Bush (R) was elected President
2008: New York 17 New England 14...?
Task 1: Use the data to make an argument for why the Democrat Obama should have won in 2008?
TAsk 2: Use the data to make an argument for why the Republican McCain should have won in 2008?
Don Fraser, OISE Professor Emeritus and humorist, likes to pose this problem to mathematics teachers at his conference talks. In effect, when you apply his two databased arguments for the year 2008, one view predicts that Obama (D) would win and the other view predicts that McCain (R) would win. It is nice that mathematics remains so consistent and "seemingly" powerful!
The "Hint" provides both of Don's arguments...but try to form your own "patterns" before looking at his.
Your new task: We now know that Obama (D) won. This means that Don's pattern continues to hold for the Democrat's side, but his "current" pattern fails for the Republican's side. You need to come up with a new datadriven argument for predicting a Republican winner using the same data but is now consistent with the new election result.
If you send me your arguments or your students' arguments, I will include them in the solution category....
Hint: Don Fraser's argument #1: Look at the point spreads for each game:
13  D
10  D
7  R
3  R
3  McCain (R) should win, as the Republican candidate always wins when the point spread is a single digit.
Don Fraser's argument #2: Find the digit sums for each team's score in a game (e.g. digit sum of 25 is 2+5=7):
10 vs 6  D
9 vs 8  D
5 vs 7  R
5 vs 11  R
8 vs 5  Obama (D) should win, as the Democrat candidate always wins when the digit sum for the winning team exceeds the digit sum for the losing team.
Solution Commentary: I know...there is nothing here....until someone sends me a datadriven argument consistent with the new data..........
And if you want to go back further, look up the Super Bowl scores for previous election years...or maybe the key is to use NCAA Final Four basketball scores....or maybe it is something connected to World Series outcomes. A lot of data to sift through...perhaps one data set is the "true" predictor!
