The predicted turnout was D+2. It was D+6. That is where things went wrong. It will be between Even and D+2 this time. Our side will win.. Gut feeling: Anti Mormon conservatives stayed home. Presbyterians don’t have that kind of bias against them.
IIRC, the turnout data was developed using the same survey methods as the voter preference data. I’m trying to remember Rasmussen, who explained it 4 years ago. I’m thinking they argued that even those calls had to be weighted where others were saying that the only discernment had to be for “registered to vote by that time” and who they were supporting. I think they ignored past voting history. They appear to have been right.
My memory is unclear on it.