Polling bias is a familiar topic, particularly among those trailing in polls.
So, you expect Republicans to howl when a Democratic polling firm produces numbers they don’t like and vice versa.
A recent example was a round of polling by a solidly Republican polling outfit, Harper Polling, which found numbers favorable to Republican U.S. Senate candidates in a number of key states (including Arkansas, with a five-point lead, but less than a majority, for challenger Tom Cotton over incumbent Democratic Sen. Mark Pryor.)
Sometimes, however, there are some solid reasons to be skeptical of polling results. Huffington Post illustrates, with an analysis of the Harper Poll in Michigan and Louisiana, It badly underrepresented the historically reliable turnout in major urban areas — Detroit and New Orleans — that have been keys to Democratic victories in those states. No similar analysis provided for the Arkansas results.
The only poll that counts, as the saying goes, is in November.