TX-Pres: Romney up 20

[krazen1211]

http://www.texastribune.org/texas-politics/2012-elections/uttt-poll-runoff-prospects-loom-us-senate-races/


Romney: 55
Obama: 35



GOP Senate candidate up 21 as well.

[krazen1211]

That's likely voters though, among registered voters it's only:

46-38 Romney

And, it's an Internet poll.

What's even more significant about Romney's 20 pt lead is the fact that McCain only carried Texas by 11 points in 2008. D/R/I of sample is 31/33/28 (R+2). The D/R/I in Texas in '08 was 33/34/33 (R+1). The partisan sample seems fair, if not slightly overly favorable to Dems.
http://s3.amazonaws.com/static.texastribune.org/media/documents/UT-TT-201205-Survey-Day1.pdf

That sample is of RV though. The release does not says what the LV sample is (probably much more Republican if the lead jumps from 8 to 20 points.

So, Obama does actually 3 points better than in 2008 with almost the same sample.

It's not almost the same sample. Obama lost by 11 in 2008 among actual (read: likely) voters, not registered voters. We don't know how much Obama lost in 2008 among registered voters.

But the poll was run by Yougov, which is of course a topnotch pollster.

[krazen1211]

How the hell do you manage to have a poll with +8 among RV and +20 among LV?  How many people are they removing from their LV screen to over double a lead?  Does anyone else think this seems insane?

Not insane, no. Rather, the respondents to such poll indicated that they were not likely to vote.

Dozens of Republicans frequently win Texas by 20 of course.