The DeSart model published an updated prediction of the election result. The model predicts that Clinton will receive 33.2% of the two-party vote share in West Virginia, while Trump will win 66.8%. In comparison, on August 7 Trump was predicted to gain 0.0% of the vote.
Putting the results in context
As any other method, econometric models are subject to bias. Therefore, as a general rule, a good strategy is to not rely too much on the results of a single econometric model. Instead of relying on results from single econometric models, the best practice is to consult combined econometric models or, even better, a combined forecast that draws upon different methods and data.
Results in comparison to other econometric models
An average of recent econometric models in West Virginia sees Trump at 60.4% of the two-party vote. This value is 6.4 percentage points lower than his corresponding numbers in the DeSart model.
Results compared to the combined PollyVote forecast
The results of the DeSart model for Trump are thus 4.9 percentage points above the combined PollyVote, which at the moment predicts a value of 61.9% in West Virginia. In comparison, a look at the PollyVote national prediction for Trump indicates that the actual results are 19.5 percentage points higher.