Notes: All polling averages are the simple mean of the last five polls from a state, combined with the Pollster.com regression line for the state. Cash on hand numbers include best estimates for matching funds.
* = Due to the dramatic, post-Thanksgiving transformation of the Republican campaign, I thought it prudent to only include post-Thanksgiving polls.
Our charts illustrate most of the variation that we can "see." Some readers may be frustrated that the lines do not shift as much as a rolling average, but more often than not, the day-do-day variation is just the "noise" of pollster house effects.
That is certainly true. Using only the latest polls to produce a polling average can skew the results based on which polling firms are included in the averages at any given moment. Then again, unless one focuses on only the most recent polls, there is a danger of missing the latest trends. I can see clear advantages to both methods
To solve this problem, for all states in the glance, I have decided to combine the Pollster.com regression line that even out “house” effects from different polling firms with the simple mean of the five latest polls from each state. To put it another way, I’ve basically decided to just average the Pollster.com and Real Clear Politics averages. Since I can’t decide which methodology I prefer, I’ll simply occupy a point of equidistance between the two and see what happens. At first look, it appears that this new methodology results in an improved outlook for Clinton.
