[bigbob]

Very intersting, thanks. So it too is a tracking poll, it was also conducted over the internet rather than on the phoone. Even more interesting that it was an “opt-in” poll and makes no claim to being a statistically-valid sample.

I think this proves my argument that probability polling is a relic of the past, with better methods now available that can better reflect the actual sentiment of voters.

Most telling is the following - another key similarity to the Daybreak poll, which only compensates for demographics, not party affiliation:

“The People’s Pundit Daily U.S. Presidential Election Daily Tracking Poll does not weight based on party affiliation (party ID). Polling results are weighted based on the U.S. Census Current Population Survey for demographics such age, sex, race, education and region”

[txhurl]

which only compensates for demographics, not party affiliation:

---

It took Trump Democrats until after Romney’s loss to really start feeling the pain zero was intentionally inflicting on the middle class. The ‘14 midterms was the best model for 2016; when 25% of Dems realized their party had been hijacked and was not likely to be rescued.

HRC 1) Refused to believe the party had jettisoned 25% of its own base, wilfully, and 2) Was too chicken to name Castro as her VP. With him, she could have achieved 8pts of vote fraud and it would be impossible to untangle or prove.

[LS]

Correct. Richard says they have used extremely detailed digging to figure out who the real “likely voters” are but that unlike the majors, they don’t go in with a “model” and try and forced respondents into a model, but analyze responses and “let the data talk to us.”