I’m now accepting anyone’s interpretation of what Ikes was asking, because it’s stuck in my head like the horse that caused me to spend that year in college
2 Likes
Yah, it’s getting bad. A “Latin American Travel Enthusiasts” Telegram group I’m in turned into all crypto/passport bros who think any Latin American country electing a remotely left leader will be the next Venezuela.
This is exactly what you’d expect if pollsters were getting increasingly Harris-favored results and reporting them as toss-ups. Coincidentally, Nate has been pounding on the desk insisting that this is happening.
1 Like
So you are agreeing that all the other polls are complete nonsense, which they’d have to be for this one to be right
It is kind of wild that the modelers out of one side of their mouth will tell you to trust the polls and throw it all into the average and don’t unskew etc etc and then out of the other say how at the end of each cycle the pollsters are either herding or we’re seeing a 1 in a mil random clustering situation
1 Like
Yeah, the Selzer poll seems especially important this time since we know for sure that other pollsters are herding.
Nate talks about it here, but the high quality pollsters have been suggesting some big racial shifts, with white people moving toward Harris and black/hispanic people moving toward Trump.
The Selzer poll – along with the Kansas/Nebraska polls that Nate cites – are imo really strong evidence that white people are indeed moving toward Harris.
We’ll see if the black/hispanic shift also manifests fully, but I personally have some doubts, especially in light of the MSG rally.
1 Like
There are two possibilities - (1) this poll is right and the other pollsters have been herding/skewing their polls to be more favorable to Trump for the reasons Nate has cited or (2) this poll is an outlier, but one of the few pollsters that has enough integrity to actually publish an outlier poll.
That said, even if it is an outlier, it’s still a pretty good sign as if you take the full MOE, it’s still only Trump +4. Even if you go to 3 standard deviations (which is 0.3%), my rough math is that it’s Trump + 8, which still is a good sign.
2 Likes
Option 3 is Iowa voters are behaving weirdly.
1 Like
Ok, white ppl shifting Harris, POC shifting Trump, whitest state in the union unexpectedly Harris
1 in 9.5 trillion!
Yeah, for the modelers it’s tough because if the polls are garbage, there’s not a lot they can do. Nate mentioned that his model does give a penalty to individual pollsters for herding, but that can’t help much if there’s a systemic problem. And if the modelers come out early and say “hey it looks like all the data is trash this year” then nobody will pay attention to their models.
Agreed. This could be the beginning of polls polling and not everyone running scared.
1 Like
Florida changed the rules so you had to request a mail in ballot instead of getting it automatically and voter turnout appears to be significantly lower. 50% of registered voters have voted vs 80% at this time in 2020. Any chance this turns into an own goal?
Seems most pollsters are cowards and aftraid to publish their real results so they all just bunch together. But the final poll is what they are judged on. This could be that we will see other polls opening up, of course maybe not.
The liberal unskew copium has always been:
- Party out of power has more energy/desire to answer polls. Libs are particularly tired after 8 years of TDS (which, for the record, is the correct response to Trump)
- Pollsters are more concerned about underestimating Trump again than vice versa, and it’s not particularly close, and therefore they put their thumb a little extra on the scale for him
Obv there’s unskew copium on the other side. Guess we’ll see
Well, if Giancarlo Sopo says it…
Also Iowa trump +5 would still be quite bad for him
9 Likes
Trump won Iowa by more than 8 in 2020, so even him being +5 now bodes well for Harris.
2 Likes