25 Oct
9:09am, 25 Oct 2024
23,294 posts
|
rf_fozzy
Trump is doing the Rogan podcast. Probably enough to tip him over the edge. I will note though that the Harris campaign isn't overly panicking and changing what they've been doing, and trump's lot have a little. It's possible the campaign polling is showing a v slightly better position for Harris than the public ones. On the other hand, trump hasn't been railing about fake polls this time around either. Maybe the internal polling of both campaigns is showing trump has already won it. And that's why he's been to California and new York. He's on a victory tour... |
25 Oct
9:27am, 25 Oct 2024
22,296 posts
|
Chrisull
Given the margain of errror on a single poll is about 3-5% (depending on what you're looking at), a 0.3% shift is nothing. Nah that's wrong. I'll quote the analyst Carl Allen : "Using Nate Silver's poll averages vs nonpartisan poll averages The fact that these guys don't understand the difference between 48.6-46.4 and 49.5-46.1 speaks to how poor their work and analysis are" I admit you strip out all the Dem polls, then you get to 49.2 - 46.2. But this is still significant. 0.6 shift closer to 50% this late on is a good sign. The reason is simple, the closer to 50% you are, the margin for error then is in your favour. works in favour of Harris. If the polls are correct, then good (doesn't guarantee an Electoral college), she's ahead but if the polls are underestimating, then she's likely above 50% and cannot be beaten. ONLY if the margin of error is below the prediction is that bad news. The probability is more likely that she will win therefore, than if she was winning say 43-39 because she could be actually on 44 and still lose if Trump is on 44-45. I'll point out same analyst says Cruz wins Texas and ignore the hype there (for the same reasons). It's still on a knife edge. But the polls flooding the zone DO make a difference. |
|
25 Oct
9:32am, 25 Oct 2024
22,297 posts
|
Chrisull
His main beef is that the polls aren't and weren't wrong, but that they are being used as a predictive tool of future performance, which they can't be used as. His argument is the analysis is bad. Polls are snapshots, and the closer you get to the end result, with the less undecideds, they become more likely to resemble the final outcome. Elsewhere I've seen some good arguments that mathematically speaking the pollsters are herding, because otherwise we'd be seeing more outlier polls, and the fact that they aren't, suggests they're tweaking ones they don't like the look of. |
25 Oct
9:34am, 25 Oct 2024
25,907 posts
|
larkim
Just trying to simplify your logic here Chris. Are you saying that if the polls split 49/46 with a MoE of 3%, then Harris has a stronger position because 50/51/52 are in play for her within the MoE, whereas Trumps MoE only maxxes him out at 49? Not sure I follow that, as surely any MoE in favour of one automatically impacts negatively with the other. I mean, clearly if the polls show any sort of a advantage, you do have that advantage in your pocket anyway; why does the proximity to 50% matter? |
|
25 Oct
9:35am, 25 Oct 2024
22,298 posts
|
Chrisull
Yes, but the polls in those states also show the story. I'll agree with you, while pointing historically, swing states all tend to go one way or the other. Similar in general elections in the UK. But yeah I admit I don't see her taking all 7. I still think, she squeaks the rustbelt, but the sunbelt goes rogue. |
|
25 Oct
9:36am, 25 Oct 2024
22,299 posts
|
Chrisull
larks - it ain't my logic, this is the analyst whose logic I used to make money on in the 2020 and 2022 elections. He was the only one predicting a senate gain for the dems in 2022.
|
|
25 Oct
9:37am, 25 Oct 2024
23,295 posts
|
rf_fozzy
Chrisull wrote: The fact that these guys don't understand the difference between 48.6-46.4 and 49.5-46.1 speaks to how poor their work and analysis are" There is statistically no difference any of those polling averages. Don't know who carl Allen is but sounds like he's taking too much copium |
|
25 Oct
9:37am, 25 Oct 2024
22,300 posts
|
Chrisull
and the margin of error positively impacting on one doesn't necessarily negatively impact on the other, because a) third party voters and b) undecideds are in the mix too.
|
|
25 Oct
9:40am, 25 Oct 2024
23,296 posts
|
rf_fozzy
larkim wrote: Just trying to simplify your logic here Chris. Are you saying that if the polls split 49/46 with a MoE of 3%, then Harris has a stronger position because 50/51/52 are in play for her within the MoE, whereas Trumps MoE only maxxes him out at 49? Not sure I follow that, as surely any MoE in favour of one automatically impacts negatively with the other. I mean, clearly if the polls show any sort of a advantage, you do have that advantage in your pocket anyway; why does the proximity to 50% matter? Actually it's 49+/- 3 and 46+/-3, so the difference is something like 3+/-5 (uncertainty isn't added linearly but in quadrature) |
25 Oct
9:40am, 25 Oct 2024
1,501 posts
|
EarlyRiser
Ignore the polls at this stage. Watch the last-leg strategies. It's all about the margins. Trump plays for the male black / hispanic vote. He won't pull much of those across. Marginal. Harris plays for the female / trad republican vote. She has more of these to win over. She'll get most of them. More than marginal. My predication stays. Solid Harris win.
|
Useful Links
FE accepts no responsibility for external links. Or anything, really.Related Threads
-
Fantasy General Election
Jul 2024
-
EU Referendum - In or Out? Vote here
Aug 2018
-
March to Parliament Against Brexit - Sat 2nd July
Jun 2016
-
EU Referendum
Feb 2016
-
Ads on Fetch - anyone else getting Leave and Remain?!
Feb 2017
-
The Environment Thread :-)
Oct 2024
-
Economics
Aug 2023
-
Dear Scottish Fetchies
Jan 2023
-
Any economists out there - question
Oct 2022
-
Power and exploitation - please check my sanity
Oct 2018
