On the night of the election, the first thing that crossed my mind was that statisticians tend to define their analyses in terms of confidence levels, or levels of significance. In other words, a .05 level of significance suggests the investigator is willing to accept a 5% failure rate with respect to his assumption (or failure to reject a null hypothesis), or that he has confidence his assertion will be true 95% of the time.
How likely is it that ALL of the swing states swung red in the same election? One would have to look at swing states and their performance in every preceding election in order to do so.
There is a simple test that determines whether a result (or score) is from the same population as other scores. Nobody mentioned in the night of the election, to my knowledge.