AZ & NV (CNN): Sinema +4 | Rosen +3

[Florida Man for Crime]

Good polls.

Now if we could only get CNN to do a GCB poll as well.

[Florida Man for Crime]

Finally, good polling on NV. Now all we need is a Fox News poll, or someone similar, of ND, and we will have good polling for every key senate race.

Erm, we need Missouri and Indiana? Only have had partisan or trash pollsters in those two recently.

[Florida Man for Crime]

CNN has an incredible big Democratic House Effect all year long. They claim Sinema leads among Early Voters 54-43....that seems to be a Non-Starter for me.

In Nevada Rosen leads Heller 49-45 among Early Voters...that seems more believable.

The NYT Siena AZ Senate poll also had Sinema up among people who had already voted, 51-45

Small samples on both, of course, but that is consistent.

https://www.nytimes.com/interactive/2018/upshot/elections-poll-azsen-3.html


While Rs do have a party registration advantage among AZ early voters, party registration is not the same as votes in a specific race, as AZ-08 showed.

In early vote turnout, Dems are in particular doing much better than 2014 among unlikely voters with bad vote history, whereas Reps seem to be cannibalizing themselves more than in 2014 in the early vote among voters who are super-likely to vote anyway (but otherwise would vote on election day). TargetSmart data on the swing in Party Regsitration of early voters as compared to 2014:

AZ --- Super Voter ---           R+3.3 ---           (D+0.7 Since Yesterday)
AZ --- Frequent Voter ---           D+1.5 ---           (D+1.1 Since Yesterday)
AZ --- Infrequent Voter ---           D+12.3 ---           (D+1.3 Since Yesterday)
AZ --- Never Voted ---           D+26.9 ---           (D+0.7 Since Yesterday)

[Florida Man for Crime]

CNN has an incredible big Democratic House Effect all year long. They claim Sinema leads among Early Voters 54-43....that seems to be a Non-Starter for me.

In Nevada Rosen leads Heller 49-45 among Early Voters...that seems more believable.

The NYT Siena AZ Senate poll also had Sinema up among people who had already voted, 51-45

Small samples on both, of course, but that is consistent.

https://www.nytimes.com/interactive/2018/upshot/elections-poll-azsen-3.html


While Rs do have a party registration advantage among AZ early voters, party registration is not the same as votes in a specific race, as AZ-08 showed.

In early vote turnout, Dems are in particular doing much better than 2014 among unlikely voters with bad vote history, whereas Reps seem to be cannibalizing themselves more than in 2014 in the early vote among voters who are super-likely to vote anyway (but otherwise would vote on election day). TargetSmart data on the swing in Party Regsitration of early voters as compared to 2014:

AZ --- Super Voter ---           R+3.3 ---           (D+0.7 Since Yesterday)
AZ --- Frequent Voter ---           D+1.5 ---           (D+1.1 Since Yesterday)
AZ --- Infrequent Voter ---           D+12.3 ---           (D+1.3 Since Yesterday)
AZ --- Never Voted ---           D+26.9 ---           (D+0.7 Since Yesterday)

TargetSmart is Trash. Also the Party ID for CNN AZ Poll is R 25 / D 25 / I 50 apparently. Sinema ain't up 11 Points among Earlies. CNN needs to get their Head examined for this sort of crap.

Party ID is not the same thing as Party Registration. I know that you are smart enough and know enough to know that is true from you making various reasonable comments to my timeline.

http://nymag.com/intelligencer/2012/11/unskewed-polls-founder-dean-chambers-on-polling-bias.html

There is no reason to be hackish, you will affect precisely 0 votes of anyone by commenting one way or another on polls on US Election Atlas, you don't have to worry about that.

I agree Sinema may well not be up among early voters (it is a small sub-sample). Likewise though, Sinema may well do better among people who don't vote early than the CNN poll or the NYT/Siena poll says.