ARG has history of poor prediction, don't take it seriously

In 2004, Gallup interviewed ARG (American Research Group) about a poor prediction (22 points adrift) in the Maryland Democratic primary and found their "likely voter" model was "based on just the one question, [which] is a relatively simple approach to classifying voters."^[4]

ARG was criticised in 2007 for only using landline respondents, which they justified by arguing that cellular phone users, who are mostly younger people, mostly "don't vote".^[5]

In the 2008 US presidential primaries, ARG were "dead last among the nine organizations that polled in 10 or more contests" for accuracy.^[11]

 

^{https://en.wikipedia.org/wiki/American_Research_Group#:~:text=American%20Research%20Group%2C%20Incorporated%20is,for%20presidential%20candidate%20John%20B.}