The DeSart model provided an updated prediction of the election outcome. The model forecasts that Clinton will obtain 50.4% of the two-party vote share in Colorado, while Trump will win 49.7%. In comparison, on August 7, Clinton was predicted to gain only 49.7% of the vote.
Historically, Colorado has been a battleground state, in which no single party has had overwhelming support to secure that state's electoral college votes. Hence, predictions in this state are of particular value.
Putting the results in context
Individual econometric models should be interpreted with caution, since they often include large biases. At the very least, one should examine how a econometric model's results compare to benchmark forecasts.
Comparison to other econometric models
An average of recent econometric models in Colorado sees Clinton at 52.1% of the two-party vote. This value is 1.7 percentage points higher than her respective numbers in the DeSart model.
The DeSart model compared with PollyVote's prediction
The results of the DeSart model for Clinton are thus 3.0 percentage points below the combined PollyVote, which at the moment predicts a value of 53.4% in Colorado. In comparison, a look at the PollyVote national prediction for Clinton indicates that the actual results are 2.3 percentage points lower.