TX SEN: Cruz +14 (CB Polling)

[The Other Castro]

It's a Google Survey poll.

[The Other Castro]

It's also worth noting that when using the "inurl:" command on Google, even the O'Rourke post shows as being published "one day ago". When I do the same for the Doug Jones post, I get an error; it hasn't been indexed by Google, which likely means it's a fresh page created within the past few hours. All other posts on the site show up with the same dates on Google as in their URL structures when queried in the same way - it's just this one post that doesn't seem to have any history!

gorelick also posted about a Delphi Analytica "poll", which, along with CB Polling, are highly likely to just be fake front groups created by some PredictIt users trying to influence pricing.

[The Other Castro]

At best, this is a group of people taking some Google Surveys and making them look pretty. At worst, it could be someone trying to use these “polls” to manipulate PredictIt markets by tricking users that don’t know any better. Regardless, CB Polling should not be considered a pollster in any sense of the word.

[The Other Castro]

Idk what Silvers’ basis is, but a history of those surveys on this forum and elsewhere should make people skeptical of the results. Anyway, CB Polling is not a pollster, nor would I be a pollster if I called myself Atlas Polling and published Google Surveys on a website. It is very easy to take multiple Google Surveys with wildly different results (as can occur with multiple Google Surveys of the same race, see MT-AL special), keep the one that’s in the best interest of the survey creator, and promote it under the guise of a legitimate polling group. I’m not saying the Google Surveys themselves are fake, just that they can be misused, and taking a Google Survey and giving your group a real-sounding name (not really though) does not make you a pollster.

[The Other Castro]

Interesting how defensive one can be regarding a group to which he or she totally isn’t connected. I did read the polls, I checked out their website, and just as they could cherrypick one out of a few surveys to promote, they could also cherrypick a particular three out of several surveys to promote. I’m not saying this kind of malicious action has been taken in this case, just that one cannot tell, and that one must be especially weary of these kinds of groups given that there have been similar groups that have done this exact kind of thing with the intent of manipulating PredictIt markets.

[The Other Castro]

There’s probably no sense in arguing since I do suspect you may be associated with this thing, but trust me I know how PredictIt works. I’ve laid out my concerns with this group and this poll.