Odd, Quinnipiac says in their methodology that it was from “self identified registered voters” but RCP seems to list it as a “likely voters” poll.
https://realclearpolitics.com/epolls/2018/senate/tx/texas_senate_cruz_vs_orourke-6310.html
ISTM, that makes a big difference in how meaningful the poll is in a place like Texas.
It makes a difference, but I don’t think it’s a big difference. Voter enthusiasm by group can shift around quite a bit in six and a half months.
Feel free to mentally shift this result a couple points in one direction or the other if you must.
I don’t get the pink shoes reference.
I would mentally shift it more than a couple points in Cruz’s direction.
It is a reference to Wendy Davis who started off fairly strong in the last Texas governor race, but faded and was beaten badly.
It is often seen that Texas democrats will put their faith in a ‘Lucy’, but then Charlie Brown runs up to kick to the football and Lucy yanks it away again. There’s a new Democratic hope in solid red Texas every couple of years, but it never pans out.
Ah, I get it. Thanks.
Go nuts.
Don’t know what you’re being pissy about but I think my point is justified. RV polls often over rate Dem support and Texas in particular has a very high Hispanic population with a particularly lower turnout rate. But if you are just blindly looking for things to cheer about, go nuts yourself I suppose.
The point is, Lance Turbo isn’t being pissy. I think you’re reading a tone of voice into his posts that he isn’t putting there. Right now, all we can really say is that there’s a chance the race could go either way, and that Cruz is probably still favored, but beyond that will depend on a whole lot of assumptions either way, and we don’t really have any good way of testing those assumptions yet. So, if you think that Cruz is doing better than 3% over O’Rourke, then that’s a reasonable guess, but if someone else thinks that O’Rourke will win by a hair, then that’s a reasonable guess, too. Myself, I know which way I hope it goes, but whether it actually will go that way, I wouldn’t even guess.
I live in TX and I agree with this, but I wouldn’t go too much more. Among the R rank and file Cruz just isn’t that popular and while Cruz will more than likely win, it will be out of inertia and not enthusiasm.
Still, I put Beto’s chances of winning at about 5%. 10% on happy days.
That’s not the point. The point is he was dismissive of the difference between RV and LV polling. I mean “feel free to move it a couple points either way”?
The difference between RV and LV is dwarfed by the massive uncertainty of extrapolating a race six and a half months from now from a single poll. That’s all.
There’s a reason why polling firms don’t do more expensive LV polls this far out.
It also has a pretty low undecided/other number for 6 months out.
IMHO the bigger issue with Texas, as opposed to a place like Alabama or Mississippi, is the absolute value of the numbers involved. Trying to make up a difference of 3% is a lot more difficult when that 3% represents several hundred thousand people rather than 10 or 20 thousand. I think O’Rourke’s chances are closer to 1% than 5% due to those numbers.
I predict that Ted Cruz is doomed.
Why would you think that?
As for RV v LV at this point in the cycle. LV screens at this point can use past voting as a predictor but most screens go beyond that and most of those questions include things that are only meaningful as an election gets closer. It is not just based on demographic weighting. They are generally not used until the election is closer not only because they are slightly more costly (need a bigger n to get your sample size) but because they are not meaningful until the election is closer.
A 3 point margin in a RV poll with a 3.6 MOE this far out, which reports that 44% of all voters have an unfavorable view of Cruz and that OTOH 53% of voters still don’t know enough about O’Rourke to know how they view him (and of those who have a view it breaks nearly 2 to 1 favorable) … means that this could be a race.
That was all I was trying to convey when I reopened this thread.
This is not directed at anyone.
What’d you go and jinx it for? Seriously, I would love for you to be right.
Texas may be a decade or two from turning purple, for a Democrat to win statewide right now would take an extraordinarily good Democratic candidate running against someone as terrible as Roy Moore on the Republican side. Cruz may not be the most likeable sort, but I can’t wrap my head around a Republican losing this state.
Actually, Ted Cruz at 47% means a very close race. Historically it seems that incumbents need to be 50 or over to be safe, 46-49 equals close race, and under 46 is near certain doom. So it’ll probably come down to how authentic O’Rourke is. Race could be another factor. Minority Republican vs. Irish Democrat makes for a rather interesting dynamic. Will racist Texans support O’Rourke? Will Cruz win more Latino votes?
Huh. I had not previously been familiar with the “50% rule” but here is a discussion (and a slight takedown) of it in an old 538 post.
Apparently over 50% in early polling does mean safe, but the converse is less true. Incumbents running under 50% in early polls still win roughly 2/3s of the time, those 41 and 47% nearly the same, and restricting to those under 45% 1/3.
The interesting thing from that post is their concluding that one should pretty much just ignore the challenger’s actual number early on.
Bottomline per that 538 analysis is that an incumbent 47% now would predict something like a final share of 53% … on average and with some decent scatter.