Sunday, October 07, 2012

University of Colorado's Election Forecasting Model Predicts a Romney Win

Well here's some additional stomach acid to keep liberals awake at night.  The University of Colorado has adopted a computer model for predicting presidential elections.  The program is based on two major inputs, (1) the electoral college votes and (2) the economic factors of individual states.  The model shows this:
According to their updated analysis, Romney is projected to receive 330 of the total 538 Electoral College votes. President Barack Obama is expected to receive 208 votes -- down five votes from their initial prediction -- and short of the 270 needed to win.
But how accurate is a computer model?  To find out, the two professors who created the model (Kenneth Bickers and Michael Berry) applied data from prior elections:
The state-by-state economic data used in their model have been available since 1980. When these data were applied retroactively to each election year, the model correctly classifies all presidential election winners, including the two years when independent candidates ran strongly: 1980 and 1992. It also correctly estimates the outcome in 2000, when Al Gore won the popular vote but George W. Bush won the election through the Electoral College.
So it appears the model is pretty accurate.  However, Bickers and Berry say it has a margin of error, i.e. an average of five states and 28 electoral votes.  It appears then, that the predicted Romney win still holds assuming the margin of error is unfavorable to the challenger.

This predicted outcome makes a lot more sense than what many polls are indicating.  In a really rotten economy, the voters generally opt to oust the incumbent and take a chance on the challenger.

Read all about it here.

No comments: