Clinton Vs. Trump: IBD/TIPP Presidential Election Tracking Poll

IBD ^ | October 28, 2017 | Investor's Business Daily

[BlueStateRightist]

The IBD/TIPP poll — a collaboration between Investor's Business Daily (IBD) and TechnoMetrica Market Intelligence (TIPP) — has been the most accurate poll in recent presidential elections.

[BlueStateRightist]

Crook up +3, D+7 sample.

[SoFloFreeper]

Not good.

[Dog]

Get off the ledge

[ScaniaBoy]

Don’t bother. It was a given, once Trump quoted the IBD/TIPP poll in a couple of his speeches.

[DoodleDawg]

The only poll that matters is a week from Tuesday.

[rb22982]

Actually their sample is R+1. They re-weighted to D+7 to get those results.

[usafa92]

Read the unweighted results. Party id is 332R 323D and 303I yet they reweighted it to D+7. This has been consistent throughout the polling period. Data is telling them that the electorate will be even yet they are reweighting it to D+7.

[BlueStateRightist]

Right. What I meant. Weighted sample of D+7 is what they’re reporting.

[woweeitsme]

I’m sorry...I must be stupid. Why wouldn’t they just run the poll at a D+7 instead of reweighting it?

[TwelveOfTwenty]

The only poll that matters is a week from Tuesday.

And we all need to show up for that one, regardless of the MSM's spin.

[Sooth2222]

"Results based on survey of 973 likely voters conducted from 10/22 - 10/27. Margin of error: +/- 3.3 percentage points. Party identification breakdown: (Unweighted) 323 Democrats/332 Republicans/303 Independents; (Weighted) 351/287/308. Results in table above are based on weighted sample. IBD's polling partner TechnoMetrica uses "traditional" telephone methodology using live interviewers for data collection for its public opinion surveys. Roughly 65% of interviews come from a cell phone sample and 35% from a Random Digit Dial (RDD) land line sample."

No polling expert but 332 -> 287 (13.6%) seems an awfully heavy Republican un-weight. Are they assuming 2008 and 2012 D-R-I voter turnout?

Live interviewers would incorporate an assumption of no "Trump Effect", too. (Do like the fact that they call 65% cell phones but even if this is one of the very best polls, I can see some problems with it.)

[KavMan]

Bill Mitchell ‏@mitchellvii 2m2 minutes ago IBD has decided to move back to the "safety of the pack" by destroying their crosstabs with ridiculous numbers. 5% w/ blacks? Come on. ============================================================ Bill Mitchell ‏@mitchellvii 37m37 minutes ago Bill Mitchell Retweeted Ned Flanders Yes, IBD has clearly jacked their crosstabs to help Hillary. From 10 down to 1 in South and 15 down to 5 w/ blacks in mere days? Bullsh*t. ============================================================ Bill Mitchell ‏@mitchellvii 49m49 minutes ago When a poll says it's been "reweighted" that means, our data doesn't meet our agenda so we "fixed" it for you.

[usafa92]

Because that wouldnt give them the result they want. Their random sample is telling them the electorate is R+1 but they either dont believe that, or in this pollsters case, is too stubborn to change his assumptions.

[Nicojones]

Because they just make the calls and that was the distribution for the period. They then re weight to get what they believe would be the actual turnout. If they kept hanging up on republicans until they got the right amount of dems (or visa Vera’s if it turned out the other way) , it would take forever to complete the poll.

[BlueStateRightist]

Most polling outfits are assuming the Crook does as well or better than 2012 Obama turnout. They are not buying into any concept of Trump monster vote or lack of enthusiasm on Dem side. In fact, they seem to be going with a concept that NeverTrumpsters are shifting the turnout against Trump relative to her lack of enthusiastic support. All assumptions that are hard to substantiate.

[Navin Johnson]

Could it be that IBD is skewing their results to please Wall Street investors and establishment types that are in the tank for Illary? Naaaah...

[ScaniaBoy]

See my post #4

[USS Alaska]

Look at the "weighted sample" that is used for the data, it is in the LAST paragraph of the article.

DemonRATs 351
pubbies 287
indies308

That is a 22% OVER SAMPLE OF DEMONRATS AND STILL IT IS ONLY A 3% BITCH LEAD.

[Sooth2222]

"Read the unweighted results. Party id is 332R 323D and 303I yet they reweighted it to D+7. This has been consistent throughout the polling period. Data is telling them that the electorate will be even yet they are reweighting it to D+7."

Are they assuming a D turnout for Clinton approximating the magnitude of the D turnout for Obama? And R turnout for Trump similar to the R turnout for McCain and Romney? Because I might be the dumbest guy in the room, but I can tell you that ain't gonna happen.

[Dog]

Good analysis