New @trafalgar_group #BattlegroundState #AZ General Election #poll (10/10-13)

47.5% @realDonaldTrump
46.0% @KamalaHarris
2.9% Other
3.7% Und

47.6% @RubenGallego
43.9% @KariLake
2.8% Other
5.6% Und#AZpol

See report: https://t.co/UulBUa3W1y pic.twitter.com/6vtxgtX9jW

— Robert C. Cahaly (@RobertCahaly) October 14, 2024

That is what I don't really understand. Gallegos is an awful candidate but Trump's coattails are stronger in other batlleground states. If Trump wins AZ by three to four points, Lake wins the senate seat, IMO. It'll be close but she wins.

If Trump pulls out a five point there, she wins more comfortably.

Same in Ohio. Trump runs up the score, Sherrod Brown goes down to Moreno. There might need 7 to 8 point win to get there as Brown is an incumbent but Trump will have coattails whereas Kamala will not.

AZ is a weird place. The McCainiacs have an iron fist on about 10% of the Republican electorate. Not enough to actually exert control against an actual conservative, but enough to play spoiler in the event of a temper tantrum. That's why Lake is struggling. McCain World hates Kamala slightly more than Trump, but they HATE Lake, and don't hate Gallego at all. If Lake wants to win, she's going to have to kiss the ring, or Trump is going to have to win AZ by 10+.

Interestingly, in August Harris was averaging 5% of self-ID'd GOP voters and Trump 5% of Dems. Today, Harris is up to 6% of Republicans, while Trump has fallen to 4% of Dems. But, Trump has largely offset that w/ slight gains among independents in recent weeks.

— Dave Wasserman (@Redistrict) October 14, 2024

Significant because Georgia insiders say Harris needs at least 90% black support in Georgia to even have a shot

Bingo.

No matter how you splice the data, Trump seems to be the strongest Republican with Black voters since 1960. Young Black men in particular have trended right during Trump's runs (cutting the Dem margin by 40 pts from 2012).

But Trump's doing historically well with Black women too pic.twitter.com/FyUqqDKLjE

— (((Harry Enten))) (@ForecasterEnten) October 14, 2024

Why the big swing?

Tom Bevan (RealClearPolitics founder) interview talking about polling, comparisons to 2020, 2016. Do we think the polls have adjusted or are they still under-counting Trump support? What can we learn about the current polls in light of what we know about past polling? 20ish minute interview.

JTN: Senate race polls begin to shift toward GOP:

Quote:

---

As the presidential race increasingly appears to be shifting towards Trump, some down-ballot races seem to be following suit, bolstering the already strong prospects of a Republican-held Senate in the next Congress.

Polling data has increasingly shown Trump gaining ground in virtually every key battleground, including Pennsylvania, Michigan, Wisconsin, Nevada, and Arizona, where key battles to decide the fate of the Senate are being waged.

Republicans are heavily favored to win the Senate, in no small part due to the layout of the electoral map this cycle. Several Democratic incumbents in swing states face reelection, notably Sens. Bob Casey, D-Pa.; Jon Tester, D-Mont.; Tammy Baldwin, D-Wis.; and Sherrod Brown, D-Ohio.

Polymarket betting odds currently give the Republicans 77% chance of taking the Senate and a 39% chance of sweeping both chambers and the presidency. Here's a look at the races pivotal to that result and shifts in each that have the bookies bullish.

---

NC poll should be good news today:

Yeah hopefully more than one. NC moved right

— Mark Mitchell, Rasmussen Reports (@Mark_R_Mitchell) October 15, 2024

This top 5 is pretty ideal for Republicans after the past 4 years:

90% of U.S. voters view the economy as extremely or very important.

Full story: https://t.co/jCy5JZhbeC pic.twitter.com/QIJWAoLF3Q

— Gallup (@Gallup) October 14, 2024

Michigan winning chances - Via Polymarket

🔴 Trump 54%
🔵 Harris 47%https://t.co/RyxUg3J9tE

— Political Polls (@PpollingNumbers) October 15, 2024

For all my reservations about Trump as the nominee again this cycle, he's really run a pretty brilliant campaign, as shown with the surge in polls he's seeing in the rust belt states (and Virginia early voting) in particular as voting starts.

We will know it's going to be a good night if Virginia is too close to call past 2 in the morning

After seeing some private polls and the latest early voting data, I’m re-upping this:

Virginia is in play. https://t.co/GSO7aQNj5g

— Election Wizard (@ElectionWiz) October 15, 2024

LMCane said:

---

We will know it's going to be a good night if Virginia is too close to call past 2 in the morning

After seeing some private polls and the latest early voting data, I’m re-upping this:

Virginia is in play. https://t.co/GSO7aQNj5g

— Election Wizard (@ElectionWiz) October 15, 2024

---

I feel like Election Wizzard had some very optimistic takes for Trump in 2020 that turned out not to be true. I cannot recall the specifics, but I'm recalling something along these lines where certain states "were in play", "safely red", etc., but none of his predictions came true.

I could be misremembering, but does anyone have opinions on the reliability of this account?


To avoid a derail: Is this site's polling/projections reliable?

Quote:

---

To avoid a derail: Is this site's polling/projections reliable?

---

my take - there is nothing reliable. but I'd rather it be trending the way Trump is than the way Kamala is.

Legend said:

---

LMCane said:

---

We will know it's going to be a good night if Virginia is too close to call past 2 in the morning

After seeing some private polls and the latest early voting data, I’m re-upping this:

Virginia is in play. https://t.co/GSO7aQNj5g

— Election Wizard (@ElectionWiz) October 15, 2024

---

I feel like Election Wizzard had some very optimistic takes for Trump in 2020 that turned out not to be true. I cannot recall the specifics, but I'm recalling something along these lines where certain states "were in play", "safely red", etc., but none of his predictions came true.

I could be misremembering, but does anyone have opinions on the reliability of this account?


To avoid a derail: Is this site's polling/projections reliable?

---

Election Wizard definitely had some optimistic red wave takes for the R's in the 2022 midterms that, as we now know, didn't exactly pan out.

This polling seems weird like that. On paper, the Dems should absolutely be in for a bloodbath. But they should've been in 2022, too. So I have no idea what to expect. I don't ever recall seeing this kind of volatility and just completely insane +/-5 point swings from poll to poll.

The video interview I linked above discusses that. The GOP actually won by several percentage points in 2022 but it didn't translate into seats due to gerrymandering and running horrible candidates in swing districts.

Another interesting point is that despite many of the polls being off in 2020 the RCP average on Election Day was actually pretty close to the outcome in many states.

Biden won Pennsylvania by almost exactly the RCP average. iirc Michigan was also close to the final RCP number. Wisconsin was the huge outlier in that Trump did way better than the RCP average.

I would not necessarily expect a huge swing on Election Day to Trump over what the polls are already saying.

People are overconfident that Trump will do better than polling like 2016. This is not a given.

📊 MICHIGAN poll by Fabrizio Ward (R) & Impact Research (D) for AARP

2-WAY
🟥 Trump: 49%
🟦 Harris: 48%
---
FULL FIELD
🟥 Trump: 46%
🟦 Harris: 46%
🟨 RFK Jr: 3%
🟩 Stein: 1%
——
MI Senate
🟦 Slotkin: 49%
🟥 Rogers: 46%
——
#140 (1.7/3.0) | 600 LV | 10/2-8 | ±4%… pic.twitter.com/glOXeN0HW4

— InteractivePolls (@IAPolls2022) October 15, 2024

📊 TEXAS GE: @hobbyschooluh

🟥 Trump: 51%
🟦 Harris: 46%
🟩 Stein: 1%
—
TX Senate
🟥 Cruz (inc): 50%
🟦 Allred: 46%
🟪 Brown: 1%

YouGov | 9/26-10-10 | 1,329 LV | ±2.69https://t.co/Lzc0FY0Eet pic.twitter.com/3z1vZJbn42

— InteractivePolls (@IAPolls2022) October 15, 2024

Harris Faulkner just showed a clip of some dem pollster guru claiming that the polls showing Trump ahead were "horse****" and that he wasn't concerned at all about Harris losing. No idea what to believe.

📊 TIPP Daily Tracking Poll

🟦 Harris: 49% [=]
🟥 Trump: 46% [=]
🟪 Other: 1%

[+/- change vs 10/11-13]
——
#115 (1.8/2.0) | 10/12-14 | 1,198 LV | ±2.9https://t.co/JWQbBQIsag pic.twitter.com/LmXMDQzQtC

— InteractivePolls (@IAPolls2022) October 15, 2024

📊 PENNSYLVANIA GE: @AmericanPulseUS

🟥 Trump: 49.0% (+1.4)
🟦 Harris: 47.6%
🟩 Stein: 1.7%

1,193 Likely voters | October 2-10https://t.co/aVfptPEF3u pic.twitter.com/lGyJr93Jw2

— InteractivePolls (@IAPolls2022) October 15, 2024

everyone pray this tweet is accurate

Older voters power Trump to Michigan lead in head-to-head race, poll reveals https://t.co/RJHof98yxc pic.twitter.com/8zN2caS1QP

— New York Post (@nypost) October 15, 2024

DonHenley said:

---

People are overconfident that Trump will do better than polling like 2016. This is not a given.

---

You're right, there are no guarantees; however, Trump is polling SIGNIFICANTLY better than 2016 and 2020 (see table below). You combine that with the past precedence of under polling Trump in 2 elections, you have to be feeling pretty optimistic if you are the Trump campaign.

aggiehawg said:

---

That is what I don't really understand. Gallegos is an awful candidate but Trump's coattails are stronger in other batlleground states. If Trump wins AZ by three to four points, Lake wins the senate seat, IMO. It'll be close but she wins.

If Trump pulls out a five point there, she wins more comfortably.

Same in Ohio. Trump runs up the score, Sherrod Brown goes down to Moreno. There might need 7 to 8 point win to get there as Brown is an incumbent but Trump will have coattails whereas Kamala will not.

---

just saw someone claims a new OHIO poll shows Trump up +11

that will win the Senate for Moreno and if true would equal a win in PA as well.

pray everyone!

Fair point.

AtticusMatlock said:

---

Another interesting point is that despite many of the polls being off in 2020 the RCP average on Election Day was actually pretty close to the outcome in many states.

Biden won Pennsylvania by almost exactly the RCP average. iirc Michigan was also close to the final RCP number. Wisconsin was the huge outlier in that Trump did way better than the RCP average.

I would not necessarily expect a huge swing on Election Day to Trump over what the polls are already saying.

---

You're accurate as to PA, but Trump over performed the RCP average in MI by 1.4 in 2020. That kind of error here on either side would be significant and possibly result changing. When we have the kind of tight numbers we have here, you have to be very precise.

JDUB08AG said:

---

DonHenley said:

---

People are overconfident that Trump will do better than polling like 2016. This is not a given.

---

You're right, there are no guarantees; however, Trump is polling SIGNIFICANTLY better than 2016 and 2020 (see table below). You combine that with the past precedence of under polling Trump in 2 elections, you have to be feeling pretty optimistic if you are the Trump campaign.

---

The Trump undercount in prior elections is certainly worth considering, but the pollsters will claim that they have adjusted based upon prior results. We shall see.

One thing that certainly does complicate this is that, in addition to switching positions on some issues, Democrats and Republicans have seen a pretty big switch in the reliability of their voter turnout. Historically, it was Republican voters that could be counted on to turn out in any election, which was one reason why they did so well in non-presidential election years in Congressional races. Now, however, the Republican base includes a lot more low propensity votes, which probably helped Democrats overperform polls in 2022.

The national party ID polls showing significant move to the Republicans in party ID are encouraging for the Republicans, but they have to actually get those folks out to vote for it to matter. That's what has to be giving pollsters a lot of heartburn. From what I've seen, while they are adjusting their samples to correspond with turnout in 2020 (e.g. if their sample is D+8, they will adjust to down to D+4 to correspond with the 2020 popular vote), they do not seem to be factoring in this recent shift in party ID (in which case, they would look for a polling same that was even, or skewed slightly R). We won't know until the results are in whether or not they are wrong to ignore that.

AtticusMatlock said:

---

Another interesting point is that despite many of the polls being off in 2020 the RCP average on Election Day was actually pretty close to the outcome in many states.

Biden won Pennsylvania by almost exactly the RCP average. iirc Michigan was also close to the final RCP number. Wisconsin was the huge outlier in that Trump did way better than the RCP average.

I would not necessarily expect a huge swing on Election Day to Trump over what the polls are already saying.

---

In both 2016 and 2020 the public polls did close a little in Trump's direction at the very last minute, and there are two explanations for this - both of which I think are true.

1) Trumps tends to be a good closer, and there was some slight to modest movement in his direction very late in both elections.

2) After having purposefully overstated Clinton's and Biden's support for the most of the election cycle, they cranked out a few more realistic polls at the very last minute so they could pretend that their polls weren't "that far off" the actual results. In essence they're saying "pay no attention to our polls from January to October when comparing them to the actual elections results... just focus on the very last one we did the weekend before the election." The mainstream media, or course, would buy it, so they would avoid being called out for having their polls serve as Democrat propaganda for most of the election cycle.

Even still... in the final polls - for both the national popular vote and the swing state polls - they still tended to understate Trump's actual vote by 4-6 points.

If we were to compare the public polls from early October, September, and August of those years, the degree to which they understated Trump's actual vote was even worse.

It is my contention in both 2016 and 2020, as well as in 2024, that Trump was never as far down as the media/university polls said... and the slightly more realistic polls they published "at the last minute" were done to preserve their "credibility" as pollsters.

txags92 said:

---

Hungry Ojos said:

---

Harris Faulkner just showed a clip of some dem pollster guru claiming that the polls showing Trump ahead were "horse****" and that he wasn't concerned at all about Harris losing. No idea what to believe.

---

I think the campaign's actions speak louder than their internal pollster's words. She would not be going on Fox News and considering going on with Joe Rogan if they were confident in a win.

---

All we have to go on really are empirical and current data. Everything else is spin.

State of the Race T minus 21 days. RCP avg is Harris +1.7 (was Biden +9.4 and Clinton +6.7). We're looking at a right shift of 5 to 8 ppts, in line with Gallup's Q3 avg of R+3. Various state polls also point to the same situation. If unchanged by 11/5, it's Trump by 312+ EV. pic.twitter.com/wHKGccbOCB

— au ng (@athein1) October 15, 2024

Quote:

---

The national party ID polls showing significant move to the Republicans in party ID are encouraging for the Republicans, but they have to actually get those folks out to vote for it to matter. That's what has to be giving pollsters a lot of heartburn. From what I've seen, while they are adjusting their samples to correspond with turnout in 2020 (e.g. if their sample is D+8, they will adjust to down to D+4 to correspond with the 2020 popular vote), they do not seem to be factoring in this recent shift in party ID (in which case, they would look for a polling same that was even, or skewed slightly R). We won't know until the results are in whether or not they are wrong to ignore that.

---

Oh I think it is prima facie wrong to ignore that large change in the electorate. Whistling past the graveyard incorrect.

When even David Plouffe is saying he doesn't even bother to look at public polls because they are BS, you know their methodologies are poor. Frankly that is because they are done on the cheap.

Mark Halperin had an older lady Dem pollster on his evening show last night. She's stopped working for the DNC an now works for Harris PACs because the money is much better. Oh and she was a loon who is in favor of abolishing the electorl college because Americans are more mobile, moving around the country and just because they happen to live in a red state shouldn't mean their "vote doesn't count."

Ag06Law said:

---

AtticusMatlock said:

---

Another interesting point is that despite many of the polls being off in 2020 the RCP average on Election Day was actually pretty close to the outcome in many states.

Biden won Pennsylvania by almost exactly the RCP average. iirc Michigan was also close to the final RCP number. Wisconsin was the huge outlier in that Trump did way better than the RCP average.

I would not necessarily expect a huge swing on Election Day to Trump over what the polls are already saying.

---

You're accurate as to PA, but Trump over performed the RCP average in MI by 1.4 in 2020. That kind of error here on either side would be significant and possibly result changing. When we have the kind of tight numbers we have here, you have to be very precise.

---

In the case of Michigan I think it's important to remember that Trump "over performed" the RCP average in 2020 ONLY if you take into account the polling published in literally the last two days of the election.

For the entire polling cycle in 2020 Biden had a huge lead in the public polling, usually in the 7-10 point range. On this day (October 15) in 2020, for example, Biden had a 7.2 lead in the RCP average, and as late as October 27, 2020 Biden had a +9 lead over Trump in the RCP average.

I think Michigan is a case of the pollsters putting out one set of more realistic numbers at the very end (in order to pretend like their polls were somewhat accurate)... after having effectively served their purpose as Democrat propaganda for most of the election cycle.