June 24 Poll Report

            Good morning everybody.  The poll yesterday was very quiet, essentially meeting expectations for all four candidates:

            That poll doesn’t move anyone’s score by more than four points.  The person most effected by it is Bullock, who drops from 91 to 87.   The removal of the old poll, the poll of May 4, was also low-impact.  That poll was Biden (71%), Bill Weld (13%), Howard Schultz (11%), and John Delaney (5%).   The removal of that data from the system helps Biden and Weld and hurts Schultz and Delaney, but not very much.   The notable changes in scores since yesterday are:

            Bill Weld is up by 11 points, and

            John Delaney is down by 6 points, and 

            Howard Schultz is down by 7 points. 

            All of those changes as a result of the removal of the May 4^th poll from the relevant data.  Weld being up 11 points (a) is only one-tenth of one percent of the vote, and (b) just returns his score to 253, which is where it was four days ago.  He controls 2½% of the market—not an insignificant amount, but not enough to make him a serious threat to President Trump, at this time.   These are the updated standings; green indicates a candidate who is up 25% in the last 30 days, and gray indicates a candidate who is down 25% in the last 30 days:

            It seems like I say almost every day that the removal of the old poll from the data has had more impact than the addition of the new one, and I have been puzzled as to why this was true.  The standings are the combined result of the last 50 polls, each of which has the same weight as the others, so adding one and subtracting one should have a balanced effect.   The "end point" polls—the first one in the series and the last one—are equally distant from the center, so that shouldn’t seem to make any difference.

            I finally realized that that happens, usually, because there was more volatility in the standings 50 days ago than there is now.   We’re kind of in the doldrums of the campaign now.  We went through a volatile stage about two months ago when new candidates were announcing their candidacy almost every day, the public was sorting them out, and a significant number of things were happening.   People were surging; people were falling off.   Everybody was finding their level.   Everybody is kind of treading water right now, waiting for the debates.  Since there was more volatility early in the early part of the 50-day window than in the latter part, the first poll in the study tends to be further from the center than the last one, even though they are equally distant in time.   Volatility should explode upward again after the debates, and then the new polls should have more impact than the old ones, I would think.   Thanks for reading.