I've seen a variety of statistical analyses of election returns purporting to suggest that there are "anomalies" that point to evidence of fraud. I've looked at a couple of these, and they all go like this.
- We will assume that the distribution of votes throughout various locations and voting methods should be the same.
- We will then analyze the differences between the various vote batches and calculate the probability that it occurred by chance alone.
- Since every statistical test to which we subject these numbers shows us that it is nearly impossible that these numbers are different due to chance alone, ergo, fraud.
Here's the problem: batches of votes should not be expected to look exactly like other batches save for random variation. One batch may be from a rural county, another from a dense urban area. One batch may be from early absentee voters, another from in person voters. These things tend to get counted together, and then reported together. These batches of votes are expected to differ tremendously from each other.
To claim that this difference is due to fraud is about the most absurd and baseless claim one could cook up--and, that claim has been submitted to the Supreme Court of the United States.
I think there is a certain type of person--ignorant of basic statistics, who hears "statistical anomaly" and automatically gets impressed. More likely, though, these claims are being made by people who know exactly how specious they are, and are being circulated by people who simply want to believe that their orange deity was cheated because that is more comforting to their ego than the idea that they lost.
While I think there's a substantial market for any idea or claim which, if believed, brings a person emotional comfort or ego protection, there's still value to understanding basic statistics. Teasing out what is random, what is not random, and why is incredibly important. Perhaps more important, though, is recognizing when one is being credulous because of one's biases or wishes.
What's Trump's next step after getting smacked down by the supreme court (for the second time...)? His instinct is to double down when he loses. This is probably when the truly dangerous antics begin.