politics and politicians
See other politics and politicians Articles
Title: Oh my: Trump 42, Cruz 32, Kasich 23 in new ARG poll of Wisconsin
Source:
HotAir
URL Source: http://hotair.com/archives/2016/04/ ... -in-new-arg-poll-of-wisconsin/
Published: Apr 5, 2016
Author: Allahpundit
Post Date: 2016-04-05 09:21:51 by Tooconservative
Keywords: None
Views: 4433
Comments: 27
We owe Trump fans some cause for hope tomorrow night. I think he’ll do better than the worst polls of Wisconsin, which currently have him down 10 points, are predicting, but, er, not this much better. Even Roger Stone thinks he’s going to lose the state, for cripes sake. This seems … sort of vaguely plausible, if unlikely? I guess? No other pollster has had Trump above 37 percent in Wisconsin, but okay. The next table, though, seems less likely. If Trump’s going to surprise in Wisconsin, you’d expect to see a big surge of independents taking advantage of Wisconsin’s open primary. That’s not what ARG sees: Trump cleans up with party regulars while Cruz holds him to a five-point lead among indies and Democrats? Highly unlikely. And for what it’s worth, it wouldn’t be unusual for ARG to miss big on a major primary during this campaign. They’ve gotten some more or less right, getting within 3.5 points of predicting Trump’s blowout win in New Hampshire and two points of his final margin in South Carolina, but they’ve had some major whiffs. Their final poll of Texas had Cruz winning by a single point. The actual margin on election day: Cruz by 17. They were the only pollsters this year to find someone besides Trump winning Michigan; they had Kasich by two just 72 hours before the vote. Trump won the state by nearly 12 points. Harry Enten, whose FiveThirtyEight site gives ARG a C- in pollster ratings, also notes that ARG somehow had Kasich a point ahead of Cruz in its final poll of South Carolina — a state where Cruz had invested considerable resources and enjoyed a natural base of southern evangelicals. Cruz more than doubled Kasich’s take in the primary, 18.5 to nine. I’m going to inch out on a limb and predict that they’ll have the order of Cruz and Trump wrong in Wisconsin too. One more bit from ARG. This may help explain why the result here is off: The lean towards Cruz there among women is probably higher than we’ll see tomorrow night but not necessarily. Multiple pollsters have showed Cruz winning big with that group in Wisconsin and one of them, Fox Business, found the same 19-point margin that ARG did. No one but no one, though, sees Cruz getting crushed by the gigantic 34-point spread among men that Trump enjoys here. To the extent that Trump leads Cruz among men in other polls, it’s only narrowly whereas Cruz leads among women decisively, which explains his overall margin. If you’re a Trump fan searching for spin and/or cause for optimism tomorrow, you’re better off skipping the ARG poll and focusing instead on the fact that Trump is poised to utterly crush Cruz in New York on April 19th. Losing badly tomorrow night is a major setback in the quest to clinch before the convention unless Trump can orchestrate a major blowout of Ted “New York values” Cruz in NY two weeks from now. Unless he underperforms in both, he’s still got a shot at 1,237. If he does underperform in both, though, hoo boy. Poster Comment: Last-minute polling, showing how widely these polls have swung this year, how few have any track record, and how often they are so completely hit-or-miss that they're kinda worthless as polls. No way to know whether this has any accuracy at all until tonight. Meaning that there's no point to it as polls go. 
Post Comment Private Reply Ignore Thread
Top • Page Up • Full Thread • Page Down • Bottom/Latest
Begin Trace Mode for Comment # 1.
#1. To: TooConservative (#0)
You're right. I think a February poll is much more accurate than one taken yesterday.
#3. To: misterwhite (#1)
They all seem to have such poor records. And the RCP running average of polls doesn't give much confidence when you realize how flaky all these polls are and how large their margins of error are. By the time you start considering polls with 5% error margins, you might as well flip a coin or just go by a state's past voting record. The state of polling is really awful, as I've said so many times. Far worse than previous cycles.
Top • Page Up • Full Thread • Page Down • Bottom/Latest"Meaning that there's no point to it as polls go."
Replies to Comment # 1.
End Trace Mode for Comment # 1.
[Home] [Headlines] [Latest Articles] [Latest Comments] [Post] [Mail] [Sign-in] [Setup] [Help] [Register]