Shillbilly Elegy Vance is yet another of a long list of examples of “stop making stupid people famous.”

The October 1st vp debate should be fun.

Harris has caught back up to her previous lead after a dip down to a lesser lead before the debate.

She has but I'm still concerned.

The issue is that she is not doing as well as Biden was at this time last year, and Biden obviously won by razor thing margines. And, because the system is fucked, a 2.0% lead is probably a loss, whereas a 3.2% lead is maybe a win.

In 2020, the polls were for the most part pretty accurate, but overestimated Biden in some states, notably Pennsylvania, Wisconsin, and Michigan. Harris' lead in those states is far thinner than Biden's was. It is possible that the models are just more accurate now and have potentially even overcorrected. There is some evidence to suggest that, such as the red wave that wasn't in 2022 and Democrats overperforming in most special elections held in that time period.

The best sign for Democrats is actually two state level polls in red states, namely Selzer in Iowa and Alaska Survey Research polls. They each found Trump winning those states by 4 and 5% points respectively. But, those are states that Trump won in 2020 by 10 and 8 points respectively. If those polls are accurate, and they're from really good pollsters, then that is a huge problem. Those states will still vote Trump, but it's unlikely that Trump would have lost so much support in Iowa while gaining support in Michigan. It could happen, but it's not likely. 

There is also the fact that Trump was actually overestimate in Georgia. And, he is doing worse now than he was in 2020. While it may seem like that was a fluke (or a steal if you're a cultist) Georgia also voted for two democratic Senators. They kept their Republican governor, but Trump has stupidly attacked him repeatedly despite his popularity. If Harris was to win North Carolina and Georgia, she would just need Pennsylvania, or any two states among Nevada, Arizona, Wisconsin, and Michigan. 

Right now Fivethirtyeight gives Harris about a 60% chance, which matches my intuition. I think it's the more likely result, but I'm far from comfortable. Which is weird when one candidate keeps talking about Hannibal Lector and black people eating dogs, but w/e.

Perhaps the polls are fairly accurate or at least a good indicator but I'm not so sure. Polling is practically the only metric I'm aware of where things are close. When you look at almost anything else like fundraising, voter registration, enthusiasm, campaign infrastructure, smaller election results, etc, I can't help but think conventional polling isn't capturing the real state of the electorate again. I could absolutely be wrong, but my gut says Harris and the Democrats outperform projections again. I'd personally give her 2:1 odds of winning and I certainly hope we hold on to the Senate and take back the House, too.

The polls have limitations, and the polls are just a snapshot. Even if they were perfect, they would tell you what would happen if the election were held today, not what will happen in November. I think they are the best tool we have available. 

The polling actually agrees with your intuition. As of now, Fivethirtyeight's model has Harris with a 64% chance. Basically the same as 2:1. It's just that 1/3 is way too big of a chance for my liking.