A few things we know right going into this election:
1: Polls have historically underestimated how well Trump does on election day.
2: Polls have historically overestimated how well candidates like Biden, and Clinton do on election day.
3: No incumbent president has ever kept the role with polling below 51% nationally (Bush, 2nd term was the lowest).
4: Both Biden and Trump enjoy approval ratings below any candidate that has won a second term (not polling but approval).
When you look at polling data, you should think “Tied is winning” for Trump. Not for his sycophants, not for MAGA, but for Trump specifically, he consistently gets more people, people who are otherwise “Not likely to vote”; a great deal of effort is placed into getting polls to be unbiased, but you can’t measure something that you don’t know about, which is fundamentally why Trump so consistently outperforms his polling. I think this also explains why progressive candidates consistently outperform their polling, but that’s beyond the scope rn.
I’m planning on digging into these for my mid-month polling update and wanted to give it a solid few weeks for the Trump trial news to saturate so we can see if there really is a significant effect. I’m also planning a ‘battleground’ state analysis for this update to start building a “road to 270” outlook.
The other thing that really matters here, is going to be the trends. There has been much ado about the latest batch of polls, but its not really clear to me that Biden has ‘ticked up’ in polling, so much as that Trump has “ticked down” post becoming a felon. To just put that into context neither Trump or Biden have been ‘improving’ in their polling since last summer, although Biden’s had dropped more precipitously than Trumps. We’ll see if Trumps ‘dump’ here is as real as the news cycle would have you believe, but keep in mind that no incumbent president has ever been competitive with an approval even remotely close to Biden’s.
My hope is by getting a little more granular into the battle ground states, the snr will be a bit better. I’m hopping to get the work done wednesday.
Polling matters, a lot. But its not a monolith.
A few things we know right going into this election:
1: Polls have historically underestimated how well Trump does on election day. 2: Polls have historically overestimated how well candidates like Biden, and Clinton do on election day. 3: No incumbent president has ever kept the role with polling below 51% nationally (Bush, 2nd term was the lowest). 4: Both Biden and Trump enjoy approval ratings below any candidate that has won a second term (not polling but approval).
When you look at polling data, you should think “Tied is winning” for Trump. Not for his sycophants, not for MAGA, but for Trump specifically, he consistently gets more people, people who are otherwise “Not likely to vote”; a great deal of effort is placed into getting polls to be unbiased, but you can’t measure something that you don’t know about, which is fundamentally why Trump so consistently outperforms his polling. I think this also explains why progressive candidates consistently outperform their polling, but that’s beyond the scope rn.
I’m planning on digging into these for my mid-month polling update and wanted to give it a solid few weeks for the Trump trial news to saturate so we can see if there really is a significant effect. I’m also planning a ‘battleground’ state analysis for this update to start building a “road to 270” outlook.
The other thing that really matters here, is going to be the trends. There has been much ado about the latest batch of polls, but its not really clear to me that Biden has ‘ticked up’ in polling, so much as that Trump has “ticked down” post becoming a felon. To just put that into context neither Trump or Biden have been ‘improving’ in their polling since last summer, although Biden’s had dropped more precipitously than Trumps. We’ll see if Trumps ‘dump’ here is as real as the news cycle would have you believe, but keep in mind that no incumbent president has ever been competitive with an approval even remotely close to Biden’s.
My hope is by getting a little more granular into the battle ground states, the snr will be a bit better. I’m hopping to get the work done wednesday.
National polls are useless. We all know full well that the national popular vote has nothing to do with who gets to be president.
yeah no. Almost all national polls are actually regional polls. So the sentiment belies a misunderstanding of how polls work.