MT-AL: Rob Quist (D) vs. Greg Gianforte (R) vs. Mark Wicks (L), May 25

[Co-Chair Bagel23]

Why the heck is this on a Thursday again?

The governor had to name a date that was 85-100 days from the vacancy.  He supposedly chose the earliest possible date, which ended up being the Thursday before Memorial Day Weekend.  He could have chosen the Tuesday after Memorial Day, but that could have caused logistical issues as well, with people on vacation.  The only other Tuesday in the period would have been June 6.

Either way, holding the election on the Thursday before Memorial Day weekend isn't ideal for turnout.

But they have had pretty ok/good ballot returns so far. I think we will see turnout in the middle to upper seventies.

[Classic Conservative]

https://mobile.twitter.com/jmartNYT/status/867152344055721985

[The Other Castro]

Pence recorded a robocall for Gianforte.

You must be logged in to read this quote.

You must be logged in to read this quote.

http://www.politico.com/story/2017/05/23/pence-montana-special-election-238743

[cinyc]

Here's a chart of all of the major Google Surveys conducted so far.  Please let me know if I missed any:

Date	Sample Size	Weighted Result	Raw Result	Included NLTV Opt-Out?
3/12 to 3/14	329	Quist+15	Quist+14	Yes
3/14 to 3/16	230	Gianforte +1	Tie	Yes
3/18 to 3/20	333	Quist +7	Quist+8	No
4/6 to 4/8	333	Gianforte +2	Quist +1	No
4/19 to 4/21	356	Gianforte +8	Gianforte +7	Yes
5/8 to 5/9	241	Quist +5	Quist +7	Yes
5/21 to 5/23	385	Quist +15	Quist +14	Yes

I reweighted all the polls using the 3-Factor Iterative Weighting (Age, Sex to the November 2014 Voter CPS and Region (59% West, 41% East)).  The link in the date column will take you to the Atlas post discussing the poll.  The link in the weighted column will take you to my weighting spreadsheets for each poll.  The link in the Raw Result column will take you to the poll itself.  Included NLTV Opt-Out means whether the "pollster" included a Not Likely To Vote option. 

[cinyc]

Can Quist really win Flathead County

If anybody can it's him. It's his home county.

This Google Poll seems to think he can - but the number of respondents from there was only 16.  That's too small a sample.  And my poll is likely crap.

If I have time tonight, I might try to look at all the Google Polls we have so far to see if the Quist-Flathead trend continues throughout.  I only started county crosstabs with this poll.

I checked.  Unfortunately, there's no reliable pattern in Google Surveys' Flathead County results. It varies.  That's probably to be expected with such low numbers in Flathead and the general suckiness of Google Surveys.

[PredictIt User]

So I ran some projections and I think Quist will start out with a early vote lead of 3-7%. I did this based off of the Governor's race in 2016, my projection showed Quist 53%, Gianforte 47%. With 10% crossover, it is Quist 54%, Gianforte 46%. Lemme know what you guys think of Quist's early vote lead (if he has one).

[PredictIt User]

NEW POLL ALERT
From Change Research/Medium.com

Gianforte- 49%
Quist- 43%
Wicks- 7%

Among those most likely to vote (NO LEAN):
Gianforte- 43%
Quist- 40%

Among those most likely to vote (LEAN):
Gianforte- 48%
Quist- 44%

Among those already decided (ALL VOTERS):
Gianforte- 42%
Quist- 38%

GOOD POLL FOR QUIST! Someone said it's an R internal too.

[Co-Chair Bagel23]

NEW POLL ALERT
From Change Research/Medium.com

Gianforte- 49%
Quist- 43%
Wicks- 7%

Among those most likely to vote (NO LEAN):
Gianforte- 43%
Quist- 40%

Among those most likely to vote (LEAN):
Gianforte- 48%
Quist- 44%

Among those already decided (ALL VOTERS):
Gianforte- 42%
Quist- 38%

GOOD POLL FOR QUIST! Someone said it's an R internal too.

If that's accurate, then crap we are screwed. Hopefully it's bad, but those are semi believable numbers, (well, besides Wicks, he is being waaaaay overpolled).

[Ronnie]

NEW POLL ALERT
From Change Research/Medium.com

Gianforte- 49%
Quist- 43%
Wicks- 7%

Among those most likely to vote (NO LEAN):
Gianforte- 43%
Quist- 40%

Among those most likely to vote (LEAN):
Gianforte- 48%
Quist- 44%

Among those already decided (ALL VOTERS):
Gianforte- 42%
Quist- 38%

GOOD POLL FOR QUIST! Someone said it's an R internal too.

Uh, source?

[Sorenroy]

NEW POLL ALERT
From Change Research/Medium.com

Gianforte- 49%
Quist- 43%
Wicks- 7%

Among those most likely to vote (NO LEAN):
Gianforte- 43%
Quist- 40%

Among those most likely to vote (LEAN):
Gianforte- 48%
Quist- 44%

Among those already decided (ALL VOTERS):
Gianforte- 42%
Quist- 38%

GOOD POLL FOR QUIST! Someone said it's an R internal too.

Uh, source?

He probably just can't post it yet. A google poll brings it up: https://medium.com/@ChngRsrch/republican-greg-gianforte-up-6-in-new-montana-poll-e1088b3a28a7

[Ronnie]

NEW POLL ALERT
From Change Research/Medium.com

Gianforte- 49%
Quist- 43%
Wicks- 7%

Among those most likely to vote (NO LEAN):
Gianforte- 43%
Quist- 40%

Among those most likely to vote (LEAN):
Gianforte- 48%
Quist- 44%

Among those already decided (ALL VOTERS):
Gianforte- 42%
Quist- 38%

GOOD POLL FOR QUIST! Someone said it's an R internal too.

Uh, source?

He probably just can't post it yet. A google poll brings it up: https://medium.com/@ChngRsrch/republican-greg-gianforte-up-6-in-new-montana-poll-e1088b3a28a7

Interesting methodology.  It's a bit unnerving that, according to this poll, a decent chuck of Bullock voters are bleeding away to Gianforte.  Quist needs to keep nearly all of them on his side to have a chance.

[cinyc]

The Change Research poll included some crosstabs with principal city info.  By my math, the sample is slightly West-heavy - but there are much fewer principal cities listed than Google provides, and it's possible some principal cities include areas in the other regions.

There is an estimated 27-point regional divide, with Western Montana going to Quist by 12 and Eastern Montana going to Gianforte by 15 among definites and probable voters, without adjusting for likelihood of voting.

Edit:  Never mind.  The votes associated with principal cities only add up to 58% of the weighted voters.  Quist actually wins them in the poll.  Without knowing where the remainder is and how it voted, we can't really calculate the divide.

[ProgressiveCanadian]

The most popular politician in the country rallies with Quist. Of course that's going to make this race close.

[Donerail]

Very weird to have a new firm emerge without even their own website, Is any more information available on them?

[○∙◄☻¥tπ[╪AV┼cVê└]

Very weird to have a new firm emerge without even their own website, Is any more information available on them?

New pollsters without websites that show up 2 days before an election are always the most reliable, don't you know?

[Shadows]

It's technically not bad, the likely voters gave a 43/40 odd to Gianforte, +3 is within the margin of error. If Quist pulls up an upset, they can claim it was always within the MOE. But it's fair to say that GOP is favored, Trump won by 20%, Dems barely squeeze in by 1-3% odd even with so-called great candidates & campaign.

What this will show is the AHCA impact - If Gianforte wins by 7-8%, then Dems have to have a multi-pronged strategy rather than relying only on AHCA - Sometimes you can't put all your eggs in 1 basket. Apart from Public lands, Quist went all in of Trumpcare with many ads, sharing stories of citizens, Trumpcare was a big part in the Bernie-Quist rallies.

I still think it is too early to say that one Trumpcare will cause a wave election in 2018 but if Quist wins or is within 1-2% odd, the GOP should be very worried for 2018 !

[cinyc]

Very weird to have a new firm emerge without even their own website, Is any more information available on them?

They have a website:
http://changeresearch.org/

I don't think there's any info on them other than what they say about themselves.  It's a new startup, polling online.  At least their poll seems plausible.  I'm not sure about their weights, though.  They seem to be giving much more weight to the youngs and too little weight to the olds compared to the 2014 CPS for actual voters.

[This account no longer in use.]

Very weird to have a new firm emerge without even their own website, Is any more information available on them?

New pollsters without websites that show up 2 days before an election are always the most reliable, don't you know?

Didn't you believe Overtime Politics?

[MT Treasurer]

It's a bit unnerving that, according to this poll, a decent chuck of Bullock voters are bleeding away to Gianforte.  Quist needs to keep nearly all of them on his side to have a chance.

This was always going to happen, though. Especially in Eastern Montana.

Anyway, regardless of who wins on Thursday, a lot of "polling firms" and people are going to have egg on their face.

[cinyc]

Change Research appears to be run by Mike Greenfield and J.R. Goldman, two Silicon Valley types.  I've linked their respective Twitter pages.  Both appear to be data scientists.  Mike Greenfield said that this is their first public poll on Twitter.

[MT Treasurer]

Change Research appears to be run by Mike Greenfield and J.R. Goldman, two Silicon Valley types.  I've linked their respective Twitter pages.  Both appear to be data scientists.  Mike Greenfield said that this is their first public poll on Twitter.

Gianforte +5 has also been my prediction since...forever now, so I hope they are right!

But seriously, why do so many "pollsters" show Wicks' support in the high single digits? The Libertarian candidate in the 2012 gubernatorial race (which was a very competitive race) won 3.7%, the Libertarian in the 2016 gubernatorial race (which was semi-competitive, I guess) won 3.4% of the vote, and I doubt they will get more than 4 in a very competitive special election.

[The world will shine with light in our nightmare]

I think that polls generally overestimate third party support.  Most of those voters usually end up breaking for either the Republican or Democratic candidate on election day.

[Shadows]

@RobQuistforMT
Thank you @MichaelKeaton for the support in this election! Montana, side with Batman—vote this Thursday, May 25!

CBO Score coming - Can that really do much for Quist? He can't really run ads at this point !

[White Trash]

@RobQuistforMT
Thank you @MichaelKeaton for the support in this election! Montana, side with Batman—vote this Thursday, May 25!

CBO Score coming - Can that really do much for Quist? He can't really run ads at this point !

This changes everything

[The Other Castro]

Article on Lake County's predicting power:

http://abcnews.go.com/Politics/single-county-watch-montanas-special-election/story?id=47570773