Nov 03, 2008 - 04:05:26 CST
I was thinking about Tuesday's presidential election and wondered if there was a way to predict which party would win the presidency based on whether or not the groundhog sees his shadow in an election year.So, I did a little research online, compiling "shadow/no shadow" stats from the Puxsutawney Phil groundhog Web site (the official site of the nation's official groundhog at www.groundhog.org/about/predictions.php) and comparing those results with presidential election results going back to 1900.
The result? In 26 out of 27 presidential election years, Phil saw his shadow. Only in 1988 did he not see his shadow.
Which means there's no correlation whatsoever between the groundhog's shadow and which party will win the presidency.
But it does mean that I can predict with a tiny bit of statistical confidence that, on Groundhog Day in the 2012 election year, Phil will most likely see his shadow.
The groundhog theory of presidential elections may be a bust, but there are dozens of other theories that have been around for a while that have varying degrees of success in predicting who will win the White House in election years.
But when it comes to polls and predictions, keep in mind the disclaimer you often hear associated with buying stocks: "Past performance does not guarantee future results."
Washington Redskins Theory
www.snopes.com/politics/ ballot/redskins.asp
Theory: How the Washington Redskins football team performs in its last home game before the election predicts who wins the presidency. If the Redskins lose, the challenging party wins. If the Redskins are triumphant, the incumbent party retains the White House.
Accuracy: In the 18 presidential elections since 1940, the theory has been correct 17 times. It was incorrect in 2004.
Prediction: To be determined. The Redskins play the Pittsburgh Steelers tonight. If the Redskins win, John McCain should win. If the Redskins lose, Barack Obama should win.
Rausch's Rule of 14
www.reason.com/news/show/34643.html
Theory: Essentially, no one gets elected president who takes more than 14 years to go from his or her first major elective office to election as either president or vice president.
Accuracy: In the 26 presidential elections since 1904, the rule has been correct in 25. The exception: Theodore Roosevelt in 1904.
Prediction: Obama was elected to the Illinois Senate in 1996 and the U.S. Senate in 2004. McCain was elected to the U.S. House in 1982 and the U.S. Senate in 1986. Obama is only four years from his first major elective office and McCain is 26 years from his first major elective office. The Rule of 14 predicts Obama should win.
Three Month Stocks
www.otcpicks.com/1121-can-stock-market-activity-predict-our-next-president.htm
Theory: The stock market trend over the three months leading up to the election will determine who wins the presidency. If stocks are up between Aug. 1 and election day, the incumbent party will win. If stocks are down between Aug. 1 and election day, the challenging party will win.
Accuracy: In the last eight presidential elections, the theory has been correct five times.
Prediction: Right now, the stock market has been trending downward since Aug. 1, which means Obama should win.
Economic Equation
http://fairmodel.econ.yale.edu/vote2008/index2.htm
Theory: Yale economist Ray Fair has, over the years, developed and refined a mathematical formula based on several economic data points that determines the percentage of the popular vote the two major parties will receive.
Accuracy: In the 23 presidential elections held since 1916, the formula has correctly picked the winner of the popular vote 21 times. It was wrong in the 1992 and 1996 elections. The results have consistently been within 2 to 3 percentage points of the actual outcomes.
Prediction: In general, based on the economic forecasts and variables used in the formula for 2008, Obama should win.
The 13 Keys
http://www.america.gov/st/elections08-english/2008/October/20081002121440adkcilerog0.7970392.html
Theory: A social formula that identifies 13 variables, or keys, that can be classified as either true or false. If five or fewer keys are false, the formula predicts the incumbent party will win. If more than five keys are false, the challenging party will win.
Accuracy: In the last 37 presidential elections since 1856, the formula has been correct 37 times. It should be noted that the formula predicts the winner of the popular vote, not the winner of the Electoral College vote. In three elections (1876, 1888 and 2000), the popular vote winner lost the electoral vote.
Prediction: The model for 2008 indicates eight keys are false, meaning Obama should win.
Pollyvote
http://www.forecastingprinciples.com/PollyVote
Theory: Combine forecasts from traditional polls, a panel of American politics experts, a prediction market and 16 quantitative models to create a single daily forecast of the popular vote percentage for the Republican candidate for president.
Accuracy: In its final prediction the Friday before election day 2004, Pollyvote correctly predicted the outcome of the presidential election and correctly predicted the Democratic Party's U.S. House of Representatives gains in the 2006 election to within a few seats of the actual total.
Prediction: While the numbers change from day to day, the trend shows McCain getting about 47 percent of the popular vote in a two-party election, meaning Obama would win.
The Unexpected
Theory: The time of greatest uncertainty on the outcome of a presidential election is when there is the greatest certainty. In 1948, John Dewey was widely expected to defeat Harry Truman. Truman won. In the 2008 presidential primaries, Hillary Clinton was widely expected to be the Democratic presidential nominee. Obama won. When compla-cency sets in, when it's assumed everything is accounted for and under control, the unexpected takes over.
Accuracy: Hard to measure since the unexpected always happens unexpectedly.
Prediction: The polls and pundits favor Obama. However, the conflicting and unsettled nature of the polls and the electorate suggest if there ever was a time for the unexpected, it would be now, which would favor McCain. The result: We'll only know for sure when the votes are counted.
(Keith Darnay is the webmaster and designer for bismarcktribune.com. His Web site, featuring this column going back to 1995, is at www.darnay.com.iec.)
Comments are reviewed for taste, tone and language before posting.
Some comments may be used in the Tribune's print edition.