I’m not sure but I’m pretty sure the answer isn’t “fuck with the dials on your model until you get to Harris +1 and ship it”.
You either trust your model and publish your results or you don’t. I get that no one wants to be the “wrong” pollster because being seen that way could be an existential threat to them. It just proves what most of us as poker players already know, that normies are really bad at understanding probability and statistics.
This is a really good point. It is one thing to complain about a singular cheating event, but to make this a foundational and persistent aspect of your political platform makes no sense at all. At a certain point voters have to ask why the Democrats are so good at this and are able to steal so many elections when they only have the support of a small percentage of Americans. At least they are able to accomplish something! Republicans have been complaining about this for a generation now and haven’t been able to do anything about it despite having the sympathy of 75%+ of the population. Why would you want to support such incompetence!
no u
A couple interesting things about the Selzer poll:
Meaning she does the poll and then adjusts for sex, age, and location. Nothing more. Kind of crazy that she’s the only one who still does it this way.
This seems good to me at first glance. The big worry is that pollsters are overtorturing the data. So this is more of a traditional, pure poll.
But, given the simple method, it’s also possible she’s missing something really important like education. She could be wrong. We’ll see.
Not including lower propensity voters is an interesting choice. Might inflate the Harris number a bit. No idea by how much.
Ask and you shall receive. Not perfect as I threw it togerher quickly, but I made one of these in 2020 and Trump lost so lets run it back.
Was reading earlier that it’s not showing up in the early voter data that Rs early voting are low propensity voters, maybe that changes on ED but it’s not showing up yet
I feel like the early vote data can be twisted whichever way suits ones narrative, because I feel like I’ve definitely seen tweets talking about the early vote has included a lot of low propensity voters.
One of things that stuck with me from Thinking, Fast and Slow is how often simple statistical algorithms beat the experts. Hopefully this is just one more example.
Do old women vote in high numbers?
They might in this election. The older women (45+) I’ve spoken to are concerned whether Harris will win but voted for her in hopes she will.
One of my older housemates surprised me by laughing at Kamala’s ads, like when Kamala says she’ll fight for the middle class, the housemate said yeah right, but she still voted for her and hopes we see our first female president.
The older women in my family, namely my mom’s sisters, who all grew up in an extremely Republican household, are having doubts about the Donald, so I take that as a pretty good sign. My Grandmother, who hates black people, is a lost cause unfortunately.
I’ve been reading for months how the dems are outpolling the republicans in the 65 and up category and I only just now realized that that population is like 55% women. Geez
Okay. Got it thank you.
And you did understand what I was going to say.
I think it’s closer to all pollsters, being
If a poll came out today showing a Trump landslide everyone would be dismissing it as a major outlier.
