MI (Trafalgar): Trump +2

[Buzz]

https://twitter.com/robertcahaly/status/1321952576884101120?s=21
Nobody else will post, so here ya go.  If they are wrong by over 4 points across the board I agree on the ban.

[Tintrlvr]

This is so obviously hot steaming trash. It was hot steaming trash in 2016, too; they just got lucky.

Reminds me of how worshipped Zogby was in 2004 after they got lucky in 2000.

[wbrocks67]

How does Trafalgar have enough $$ to constantly poll these races so much?

[Darthpi – Anti-Florida Activist]

This is just funny at this point.

[Darthpi – Anti-Florida Activist]

How does Trafalgar have enough $$ to constantly poll these races so much?

Doesn't take much money to just make up the data out of thin air.

[WD]

lmao the link to their poll is dead

[Rand]

Watch them be right again 

[Middle-aged Europe]

Thanks Trollfalgar.

[Badger]

This is just funny at this point.

[Gass3268]

This is the exact same margin as their disputed poll from over the weekend, I'm not even sure if the decimals changed.

[Hammy]

This is the exact same margin as their disputed poll from over the weekend, I'm not even sure if the decimals changed.

Makes me feel better about ignoring it. Especially as the link is broken

[n1240]

lmao the link to their poll is dead

It usually takes them a few hours to actually upload it to their site. Looking forward to viewing the nonsensical crosstabs once again though!

[User441]

As Red Letter Media would call it, these guys are hack frauds.

[Illiniwek]

Cancel the election

[Wiz in Wis]

Guys and Gals... this seems like hot steaming trash:

https://twitter.com/JMilesColeman/status/1321993837875789824

[Senator-elect Spark]

https://twitter.com/robertcahaly/status/1321952576884101120?s=21
Nobody else will post, so here ya go.  If they are wrong by over 4 points across the board I agree on the ban.

No way

[lfromnj]

Guys and Gals... this seems like hot steaming trash:

https://twitter.com/JMilesColeman/status/1321993837875789824

I mean thats like 1/13 of the state electorate right?

MOE especially when you aren't even weighting by CD can be absurdly high. Selzer has pretty garbage results for its CD polls too.

[Wiz in Wis]

Guys and Gals... this seems like hot steaming trash:

https://twitter.com/JMilesColeman/status/1321993837875789824

I mean thats like 1/13 of the state electorate right?

MOE especially when you aren't even weighting by CD can be absurdly high. Selzer has pretty garbage results for its CD polls too.

1) There is difference between 4 CDs and 13 CDs,
2) The demos (age/race/party) are also really really bad. This has Trump winning 60% of voters under 25, 25% of blacks... 1/3 of Democrats!

[Tintrlvr]

Guys and Gals... this seems like hot steaming trash:

https://twitter.com/JMilesColeman/status/1321993837875789824

I mean thats like 1/13 of the state electorate right?

MOE especially when you aren't even weighting by CD can be absurdly high. Selzer has pretty garbage results for its CD polls too.

It's a way that their results can be shown to be fraudulent, though. For example, they say their N is 1058 LVs, and 5.2% are in MI-13. That would mean 55 voters in MI-13. So far, so good - the numbers are coming out to integers (assuming some rounding in the percentages).

Yet the percentages of the vote in MI-13 don't give you integer numbers of voters when taken to be of 55 voters. For example, they say Biden has 55.7% of the vote in MI-13. 55.7% of 55 voters in MI-13 is... 30.635 voters? No way that's due to rounding; they just made up the numbers. Note that they don't list out the number of voters in the crosstabs, because the crosstabs are fictional.

Another example: They say their sample is 1.8% Asian. That's 19 voters. Yet somehow 2.7% of those 19 Asian voters "lean Biden" (that's 0.513 voters), and another 3.5% "lean Trump" (that's 0.665 voters). Pure fiction!

[Amenhotep Bakari-Sellers]

Speaker McCarthy likes to say hi, we know these polls are only propping up Trump from Fox News. The arm of Russian and Iran Eric and Don Jr.

We now know that Reagan was a Russian Prez too since the Iran Contra situation and Iran in a Russian state. That's why in elementary school we had nuke weapon drills in case Russia hit us. Russia, Iran and China control oil

[lfromnj]

Guys and Gals... this seems like hot steaming trash:

https://twitter.com/JMilesColeman/status/1321993837875789824

I mean thats like 1/13 of the state electorate right?

MOE especially when you aren't even weighting by CD can be absurdly high. Selzer has pretty garbage results for its CD polls too.

1) There is difference between 4 CDs and 13 CDs,
2) The demos (age/race/party) are also really really bad. This has Trump winning 60% of voters under 25, 25% of blacks... 1/3 of Democrats!

The 2nd is a legitimate gripe, however the first point means its even harder to get accurate CD polls with more districts assuming the same total sample size. Heck a 1000 n CA poll has a small chance of showing Maxine Waters losing.

[Amenhotep Bakari-Sellers]

Rule of thumb in core three states if QU or PPP says D's are winning WI, MI and PA🎃🎃🎃, D's are winning them, QU just polled them Biden is the 47the Prez Wed

[n1240]

Tiny sample CD tabs can be very noisy but there is really no justification for a poll showing 25-30% of both Dems and Reps crossing over, and Trump winning independents by 28% should while winning more Democrats than Republicans should translate to a double digit lead for him, not a two point lead, unless the sample is like 50/30/20 D/R/I for whatever reason.

[Admiral Stockdale]

I would think if you got these types of results that you would question it.  If they are true/accurate then I guess just like Trump voters fake out pollsters, maybe Biden voters can too...

• 18-24:  58.2% for Trump
• 25-34:  45.6% for Trump
• 65+:    41.8% for Trump
• Independents:  59.3% for Trump
• Democrats:  29.0% for Trump

[Senator Incitatus]

I predict the next Trafalgar poll, wherever it is, will be within 3 points.