Sabitlenmiş Tweet
FINAL PollFair Presidential Election Forecast @270toWin
🔴Trump 297 ✅
🔵Harris 241
#GeneralElection #2024Elections #Trump #Harris #President #Map
English
PollFair
1.7K posts
PollFair
@poll_fair
Ensuring fair and accurate political polls by reweighting polls based on historical exit poll data.
Katılım Ağustos 2024
51 Takip Edilen3.3K Takipçiler
@poll_fair There is a sizable group that wants a Trump third term. I think that explains this.
English
Which is more likely?
That the country somehow moved 12 points to the left in a few months, a signal picked up by no one?
Or that the "most accurate pollster" isn't actually accurate.
English
All AtlasIntel had to say was it's possible they had a bad sample and it could be a fluke, but they decided to triple down and now it's looking more horrible by the minute
Mark Mitchell, Rasmussen Reports@honestpollster
Trump is not above water in a SINGLE issue in the Atlas poll. Ok.
English
@EricLDaugh @atlas_intel He still gave high ratings to many pollsters that called almost every swing state wrong.
English
🚨 #BREAKING: Nate Silver marks ATLAS INTEL as the top, A+ rated pollster, after nailing Donald Trump's victory over Kamala Harris in 2024.
Congratulations, @Atlas_Intel.
English
@poll_fair Thank you for your excellent service and analysis
English
It's looking like our election day model was pretty accurate, except for the Florida senate race which we called correctly but the margin was off.
Florida
PollFair Call: Trump +11
Actual Result: Trump +13
PollFair Call: Scott +8
Actual Result: Scott +13
Nevada
PollFair Call: Trump +4
Actual Result: Trump +4
PollFair Call: Rosen +3
Actual Result: Rosen +2
Arizona (90% in)
PollFair Call: Trump +6
Actual Result: Trump +7
PollFair Call: Gallego +1
Actual Result: Gallego +2
Our election day model is partially based on average polling crosstabs, providing more evidence that the poll surveys themselves were actually quite accurate and just the original poll weightings were wrong in many cases.
PollFair@poll_fair
Based on election day electorate so far, and electorate of the early and mail-in vote, we're currently projecting: Florida: 🔴Trump +11 🔴Scott +8 Arizona: 🔴Trump +6 🔵Gallego +1 Nevada: 🔴Trump +4 🔵Rosen +3 (Based on the 30-day rolling average of state party id crosstabs)
English
Based on election day electorate so far, and electorate of the early and mail-in vote, we're currently projecting:
Florida:
🔴Trump +12
🔴Scott +8
Arizona:
🔴Trump +7
Senate TIED
Nevada:
🔴Trump +4
🔵Rosen +3
(Based on the 30-day rolling average of state party id crosstabs)
English
Pre-election favorability ratings suffered from the same weighting issues that caused polls to be inaccurate in calling the election itself, which caused Harris/Walz to appear liked and Trump/Vance to appear disliked. You can see closer to reality in the exit polls.
Political Election Projections@tencor_7144
CNN National Exit Polls: Favorability Ratings Donald Trump: 46/53 Kamala Harris: 47/52 Tim Walz: 45/47 JD Vance: 47/46
English
FINAL PollFair Presidential Election Forecast @270toWin
🔴Trump 297 ✅
🔵Harris 241
#GeneralElection #2024Elections #Trump #Harris #President #Map
English
@TheXclamationX We thought they were only counting mail in ballots at this point, which is why it has swung in Whitesides favor, but we could be wrong.
English
@PerfectUnion87 Good question. From what we've seen some devise a weighting based on various dynamics, others use the survey itself to influence the weighting, or a combination of the two. It is certainly open to mistakes and/or abuse.
English
It looks like the average weighting from national polls ended up being just a fraction above D+0. Preliminary results from exit polls suggest the real environment was R+3.
Toward the end, we saw the more accurate pollsters homing in on R+2, but during the whole campaign we saw anything ranging from R+2 to D+7. We find it hard to believe any pollster truly thought D+7 would be accurate.
The interesting thing is the party-id crosstabs in most polls were close to accurate - if you apply an R+3 weighting you generally get results ranging from a tie to Trump +3 nationally from most polls, and our average would be Trump +2.
So, there was nothing wrong with the surveys themselves in the main - it was almost entirely a weighting issue, as we suspected.
English
@poll_fair Great work this year.
When you eventually build a website, it’d be interesting to add sliders where a user can change the electorate to whatever environment they’d like and see how all of the toplines are affected for each poll.
English
@poll_fair hilarious. I love remarks like: "I live in XXX place therefore I know how wrong this poll is". I saw that over and over.
English
This was our most viewed post of the election cycle for whatever reason. We got a lot of flak for what we are doing, but ironically it looks like this Fairweighted poll might end up being dead on accurate - Baldwin +1. Read the comments if you need some entertainment this morning!
PollFair@poll_fair
2024 Wisconsin Senate GE - Siena (NY Times), 9/21-9/26, 688LV Fairweighted™ Poll: 🔴 Hovde 47% 🔵 Baldwin 48% (+1) PollFair Weighting: R+5 (32% D, 37% R, 31% I) Siena Weighting: D+1 (31% D, 30% R, 38% I) Original Poll: Hovde 43% Baldwin 50% (+7) #WisconsinElection2024 #FairPolling #WIPolitics #WI2024Elections #ElectionIntegrity #WisconsinPolls #PoliticalAnalysis #WIGeneralElection #PollAccuracy #ElectionData #Hovde #Baldwin #Republican #Democrat #Senate
English
@poll_fair @USA_Polling Well clearly you weren’t unskewing polls because you got it right.
GIF
English
They're fucking unskewing the polls again!!!!!!!!!!!!!!!!!!!!
PollFair@poll_fair
2024 Wisconsin Senate GE - Siena (NY Times), 9/21-9/26, 688LV Fairweighted™ Poll: 🔴 Hovde 47% 🔵 Baldwin 48% (+1) PollFair Weighting: R+5 (32% D, 37% R, 31% I) Siena Weighting: D+1 (31% D, 30% R, 38% I) Original Poll: Hovde 43% Baldwin 50% (+7) #WisconsinElection2024 #FairPolling #WIPolitics #WI2024Elections #ElectionIntegrity #WisconsinPolls #PoliticalAnalysis #WIGeneralElection #PollAccuracy #ElectionData #Hovde #Baldwin #Republican #Democrat #Senate
English
@chompyfone We were very confident we'd turn out to be more accurate than most polls, even though we ended up being too far left in some cases. And that ended up being true.
English