Should the next person who posts the "Regression toward the mean'' wiki link...

[SPC]

I would think that individuals who subscribe to such pseudoscientific beliefs as thinking that a pollster showing numbers consistent with the consensus after previously showing numbers dismissed as an outlier in favor of the leading candidate somehow constitutes "momentum" for the trailing candidate, or that a sample size of two cherry-picked recent elections in which disreputable polling firms constituted a disproportionate share of the polling average in certain states provides evidence of "overpolling," or that analyzing demographic crosstabs with a much smaller sample size than the main poll provides a more accurate picture of the state of a race than the poll itself, would be more deserving of such treatment, rather than those who seek to inculcate a basic sense of statistical knowledge in their fellow poll watchers.