Today delegates will go to the polls in Utah to determine who will represent their party. PoliticIt has applied the It Score algorithm to find out who will likely be the representatives.
The It Score is a measure of a digital influence that correlates with election results (GigaOM).
The score was used in the 2012 election to correctly predict the outcomes of every major federal race with 92% accuracy, and it correctly predicted every race in the 2012 Utah conventions.
If a candidate has a higher It Score relative to their opponent then they will likely win.
Scores for candidates running against each other are calculated on a relative basis. Methodology for its calculation is contained at this end of this article.
Utah Convention Election Forecast
Rob Bishop, Chris Stewart, and Jason Chaffetz all have challengers, but their opponents lack sufficient data to calculate an accurate score. Each of these incumbents will likely win today with big margins.
There are really only three races to watch. The first is between democratic candidates Donna McAleer, and Peter Clemens in district 1.
McAleer has an It Score of 50.37, and Clemens has a score of 49.62. McAleer has a slight edge over Clemens, but this race could really go either way.
Another race to watch is between republican candidates Mia Love, and Bob Fuehr in district 4. Love has an It Score of 91 compared to Fuehr who has a score of 9. It should be noted, however, that most of Love’s buzz is coming from outside the state of Utah. Love will likely win today.
Doug Owens, and Bill Peterson are competing for the democratic nomination in district 4. Peterson has almost no digital presence making it impossible to generate an accurate score for him. Owens will likely win today. Owens isn’t very strong relative to Fuehr and Love…he would do well investing in a stronger digital presence prior to the general election in November.
Those are the numbers for this years convention. What do you think?
The It Score is a machine learning algorithm that gathers chatter surrounding a political candidate from social, and traditional media sources in order to provide a gauge of their digital influence. It accounts for tone, how people are reacting to a politician, the buzz surrounding the candidate, and what people are saying about the politician.
Whatever candidate has the highest It Score will likely win in the election.
The machine learning algorithm was trained off of actual primary election results. The original algorithm was predicting at 67% accuracy, but overtime it learned. Currently it is predicting at 92% accuracy, and has a standard error of 5.
One limitation the algorithm faces is it’s inability to predict races where candidates lack a presence online. PoliticIt omits these races because there is insufficient data.
PoliticIt’s hope is to refine this algorithm so that political candidates can use it to receive real-time feedback on campaign performance.
(Photo Credit: Flickr via Elizabeth Foote)