MS-Change Research: Hyde-Smith +5

[ON Progressive]

Cindy Hyde-Smith (R-inc) 51
Mike Espy (D) 46

https://twitter.com/ChangePolls/status/1067122676819406848

[ElectionsGuy]

Change Research is terrible, they had Rosendale winning by 3 and Casey (PA) doing worse than Brown (OH). Their methodology is terrible.

[MT Treasurer]

Looks about right. Hyde-Smith is probably favored, but Democrats have a very high floor in this state, so if black turnout is through the roof/enough whites stay home...

[ON Progressive]

Change Research is terrible, they had Rosendale winning by 3 and Casey (PA) doing worse than Brown (OH). Their methodology is terrible.

Agreed. They also had McDaniel winning second place (LOL) in November.

[Tender Branson]

Change Research is pretty bad.

Still sticking with my Cindy+11 prediction (+/-2%).

... but who knows what happens after my disaster predictions for Senate and Governors on Nov. 6th.

[Matty]

Can someone explain to me how a poll of 1300 people randomly distributed across Mississippi can be junk?

Doesn't statistical theory say that is more than enough of a sample size to be accurate?

[Comrade Funk]

Can someone explain to me how a poll of 1300 people randomly distributed across Mississippi can be junk?

Doesn't statistical theory say that is more than enough of a sample size to be accurate?

The pollster was pretty bad during the elections.

[SATW]

Can someone explain to me how a poll of 1300 people randomly distributed across Mississippi can be junk?

Doesn't statistical theory say that is more than enough of a sample size to be accurate?

Do you not know how to read? Other posters listed things that could be wrong with the poll.

[BundouYMB]

Change Research had McDaniel ahead of Hyde-Smith in a poll released on November 4. Take it with a grain of salt (and then another couple grains, or possibly a whole tin of salt.)

[Amenhotep Bakari-Sellers]

Hopefully, Espy can do what Bredesen couldn't do, win in a red state

[History505]

Don't believe it.

[Amenhotep Bakari-Sellers]

51/49 my prediction Espy.

[IceSpear]

They had Wicker winning by 8 a day before the election. He won by 19. It's literally on the first page of this board.

https://uselectionatlas.org/FORUM/index.php?topic=305958.0

[Amenhotep Bakari-Sellers]

If Espy upsets, I would be one of the few ones that predicted an Espy win, due to fact, the election predictions are closed for updates.

[Beet]

I choose to believe this poll.

[Xing]

This is probably the best case scenario for Epsy, but I think he'll lose by more like 10-12%.

[Esteemed Jimmy]

Don't believe it.

[RussFeingoldWasRobbed]

J
U
N
K

[Senate Minority Leader Lord Voldemort]

Titanium R.

[cinyc]

Can someone explain to me how a poll of 1300 people randomly distributed across Mississippi can be junk?

Doesn't statistical theory say that is more than enough of a sample size to be accurate?

You could poll 10,000 Mississippians. If your methodology stinks or weighting is off, it won't matter one bit that the sample size is larger.

A one-question Google Survey with a sample size of 10,000, for example, would still be crap because of bad methodology.

[History505]

Hopefully, Espy can do what Bredesen couldn't do, win in a red state

Trump's two rallies for Hyde-Smith is going to energize his base to come out.

[I Can Now Die Happy]

Day before the election, I noticed that Change Research accurately predicted Baria's and Espy's percentage on November 6, but not the GOP's, so I figured if that same pattern held, that Espy would get 46%, but that Hyde-Smith would get the remaining 3%, leading to her getting 54%.

and it seems like that's what the final result is going to be. Should have bet on that instead of trying to extrapolate from election day results (on predictit, the winning bet is either going to be the 6-8% bracket or the 8-10% bracket; I originally bet 12-14%).

[Zaybay]

Eh, looks like the junk poll was pretty close.