Posted by Sarah John on November 11, 2016
In what many might have been assumed to be an unpredictable election, FairVote was able to project the outcomes of over 80% of U.S. House seats with 100% accuracy two years ahead of Election Day 2016.
In November 2014, two days after the November 4th midterm elections, FairVote made its final 2016 US House projections as part the biennial Monopoly Politics report. FairVote projected the outcome of 362 of the 435 House races: 202 to Republican Party candidates and 160 to Democratic Party candidates. The only subsequent changes were formula-driven ones if a seat became open or, as in the case of a few states, district lines were redrawn.
While provisional and mail ballots are still being counted, and the winner in Louisiana’s fourth congressional district will be decided by runoff next month, FairVote can confidently say that we were 100% accurate in our projections of 2016 US House races.
This accuracy was achieved despite the fact that we made our projections in 2014 based only on the 2012 and 2014 election results in each district and the presence of an incumbent. We did not use campaign funding data or adjust based on opinion polls. The only information someone needs to predict with high accuracy the outcome of most U.S. House elections is previous presidential and U.S. House election results.
We also make full projections of every seat in the U.S. House, including those that should be “too close to call” or genuine “tossups”. Once again, using only previously election results, we were 97% accurate in our projections while only factoring in election information from 2014 and 2012.
Using the same methodology, FairVote will project outcomes for the November 2018 election once data is available from presidential results by congressional district Stay tuned.