Just when pundits were rushing to explain why Biden was doing poorly in PA, we suddenly get two polls with big leads. Now the PA poll averages look good for him
A mathematician friend tells a story of how he gave a lecture to a large audience. He had everyone flip a coin, and then ask their neighbors what they got. The audience members were then asked to report any strange patterns. One group excited said they had a cluster all heads...
...the same goes with state polling. With so many polls, it is possible that by chance alone we'll see an odd sequence of random draws. Are the two PA polls outliers in the same direction? Are the other polls outliers?
Poll averaging helps to a certain extent, but it can fail when by sheer bad luck two or three polls in a row are outliers in the same direction. And since the horserace analysis is time-dependent, pundits have a tendency to over-analyze recent polls if they show "tightening"
So, last week's punditry of "Biden doing poorly in Appalachia, costing him PA" replaced two week's ago punditry that "Biden is doing poorly among FL Hispanics, costing him FL". Who knows where the next set of correlated outliers will be, but I guarantee they will happen
Share this Scrolly Tale with your friends.
A Scrolly Tale is a new way to read Twitter threads with a more visually immersive experience.
Discover more beautiful Scrolly Tales like this.