Polling by right-leaning firms has exploded this cycle. Maybe they want to be accurate—or maybe they’re trying to create a sense of momentum for Donald Trump.
I fully believe that, but I also know the polls have been absolutely useless since 2016, and P2025 and Rowe are such massive issues, I expect the actual exit polls to be heavily in favor of Democrats.
The (independent) polls were generally good in 2022, it was the polls with a bias and the big aggregators who totally missed. Several of the aggregators who ignore clearly biased polls called a few races, like Fetterman, with high accuracy.
This time around they show Harris with a 0.5-2pt margin in PA, MI, WI, and NV. Trump with that same margin in NC, GA, AZ. I think that means that Harris is favored for the EC, but that we need turnout.
Look at all the post-Roe states where abortion was on the ballot. Landslides for the amendments, and a +10 points minimum improvement for Dem candidates. I think Kansas even had a +20 swing towards Dem candidates, but I can find the exact number.
Polls have not been able to adjust to modern technology, and have never taken into account incoming new voters, the largest group in the country since 2000. Every election since 2016 specifically has blown out populous voting records.
You can’t rely on technology affluent people to get accurate internet, cold calling, or questionnaire based answers. Polls are shit.
Harris has this. I’ll come back here and dance for you if I’m wrong.
I think we might actually agree more than you imagine - I also think Harris is doing pretty damn amazing right now, and I also think it looks good for her.
I’m not suggesting the forecasts for things like the amendments were correct, they definitely missed, and hard. I’m saying they were wrong because they took in a lot of clearly biased inputs.
There were other polls that actually had a lot of this data in them, and showed a clear lean in the odds post-Roe. However, these polls were being weighted by aggregators against stuff like Rasmussen, and Trafalgar, which are absolute trash. The forecasters were applying weights they themselves invented to these polls and including the trash data, meaning it was trash data AND it was deliberately turned into something that biased the sample set towards a middle average.
What I’m saying is that cutting that chaff out of the results, and then being realistic about what a “+2 margin” means (it’s actually pretty good) results in a wholly different picture than the aggregators are giving us. One where Harris is more or less the clear pick.
Anyway, having said all that - it really, really does come down to turnout on this one. Trump’s base doesn’t really falter, and it’s around 65m votes every time. That can get flooded out but not without people showing up.
You should be, because there’s more than a 0% chance it happens, and when it does, it’s likely the end of America as we know it.
Not enough people know that it really would be the end.
I fully believe that, but I also know the polls have been absolutely useless since 2016, and P2025 and Rowe are such massive issues, I expect the actual exit polls to be heavily in favor of Democrats.
The (independent) polls were generally good in 2022, it was the polls with a bias and the big aggregators who totally missed. Several of the aggregators who ignore clearly biased polls called a few races, like Fetterman, with high accuracy.
This time around they show Harris with a 0.5-2pt margin in PA, MI, WI, and NV. Trump with that same margin in NC, GA, AZ. I think that means that Harris is favored for the EC, but that we need turnout.
Uh, no they were not. Not even close.
Look at all the post-Roe states where abortion was on the ballot. Landslides for the amendments, and a +10 points minimum improvement for Dem candidates. I think Kansas even had a +20 swing towards Dem candidates, but I can find the exact number.
Polls have not been able to adjust to modern technology, and have never taken into account incoming new voters, the largest group in the country since 2000. Every election since 2016 specifically has blown out populous voting records.
You can’t rely on technology affluent people to get accurate internet, cold calling, or questionnaire based answers. Polls are shit.
Harris has this. I’ll come back here and dance for you if I’m wrong.
I think we might actually agree more than you imagine - I also think Harris is doing pretty damn amazing right now, and I also think it looks good for her.
I’m not suggesting the forecasts for things like the amendments were correct, they definitely missed, and hard. I’m saying they were wrong because they took in a lot of clearly biased inputs.
There were other polls that actually had a lot of this data in them, and showed a clear lean in the odds post-Roe. However, these polls were being weighted by aggregators against stuff like Rasmussen, and Trafalgar, which are absolute trash. The forecasters were applying weights they themselves invented to these polls and including the trash data, meaning it was trash data AND it was deliberately turned into something that biased the sample set towards a middle average.
What I’m saying is that cutting that chaff out of the results, and then being realistic about what a “+2 margin” means (it’s actually pretty good) results in a wholly different picture than the aggregators are giving us. One where Harris is more or less the clear pick.
Anyway, having said all that - it really, really does come down to turnout on this one. Trump’s base doesn’t really falter, and it’s around 65m votes every time. That can get flooded out but not without people showing up.
It’s going to skew way harder for Harris than +2. Sit tight.