Related questions
@ShadowyZephyr It won't go blue in 2024, but it has been noticeably narrowing up throughout the 21st century, and a Democrat won a statewide race in 2022. That makes it least plausible that a Democrat could win it sometime in the not-too-distant future.
@ShadowyZephyr I don't disagree with you there. The main reason for my comment is that, even though Alaska has gone red for the last 14 elections, that alone isn't a good reason to think it will go red in the next one. If Alaska had always gone red but was consistently getting narrower, and the previous margin was 1%, then that would be a good reason to expect it to finally flip.
The main reason it's very unlikely for Alaska to go blue in the next election is because it hasn't been close in any recent elections. Even in 2020, which was the closest, Trump still won it by 10 points, and states very rarely change their margins by >10% in a single election.
@JosephNoonan It's all about the margins, yeah - e.g. Wisconsin was blue in every presidential election from 1988 to 2012, then flipped in 2016 because it was always narrowly blue. Similarly, Georgia was red for a long while until 2020.
For reference, the number of states that have had margin shifts of at least 10% in recent presidential elections:
2020: 0 (Vermont and Colorado were the closest, at roughly D+9% and D+8%)
2016: 9
2012: 5
2008: 27 (26 towards Democrats, 1 towards Republicans)
2004: 3
2000: 23 (all towards Republicans)
2000 and 2008 represented huge shifts in the national popular vote, so let's discount those and just calculate stuff using data from 2004, 2012, 2016, and 2020:
Base rate of at least one state swinging by 10%: 75%
Average number of states that swing by 10%, if any do: 17/3 = 5.67
5.67/51 (states + DC) = 11% chance of a state swinging that much, if the chance is equal for all states
The chance is not equal; small, rural states tend to be swingier than large, urban ones. 10 times out of 17, the states that swung by 10% or more were relatively rural states with relatively small populations (i.e. 5 or fewer congressional seats). Of the states that swing that much, we would expect 58% to be small and rural, or 3.29 total.
18 states meet my criteria for "small and rural," and 33 (including DC) do not. 18/51 = 35%; these states make up 35% of all states but 58% of states that swing by 10% or more, which is a large enough disparity for me to be confident that there is a correlation.
Let's simplify and say 3 states of the 18 that meet my criteria will swing by 10% or more in either direction - that would mean each small, rural state has a 1/6 chance to swing by 10 points or more, or 16.67%.
25% chance no state swings by 10% or more and a 16.67% chance Alaska swings by 10% or more, given any state does (i.e. 83.33% chance it doesn't if at least one other state does, so 83.33%*75%) = 100% - 25% - 62.5% = 12.5% chance Alaska in particular swings by 10% or more in either direction in 2024, assuming it is as likely to do so as any other relatively rural state with a relatively small population.
Add in the fact that Alaska has been shifting to the left at an accelerating pace for more than a decade now (meaning a shift to the right by 10% is very unlikely), plus uncertainty from being a year and a half away from the election (weird stuff happens in politics all the time; if you told someone in May 2019 that Biden would win Georgia but lose North Carolina and Florida, they'd probably look at you like you were insane), and Mary Peltola probably helping Democrats' image in Alaska and campaigning for Biden, and that's why I keep betting the market up to around 15%. I don't think Biden will win Alaska, but I think the chance is probably closer to one in six than one in ten.
@JosephNoonan Yeah I agree with you. I just put that there to inform anyone who clicks on this market and didn't know that. The margin was a lot.
@evergreenemily I feel like you are missing how unpopular Biden is as a candidate right now. given the high chance of him being the nominee, that lowers my estimate for the chance that Alaska will swing. Also, the election cycles that have the higher amount of states swinging are ones where the candidates are different from the previous ones. Races in which a candidate runs for re-election don't seem to swing nearly as much. (2004, 2012, 2020).
From 2016 to 2020, the blue swing of Alaska was not much more pronounced than the rest of the nation's (4% vs 2%.) Same thing from 2012 to 2016: (2% vs 0%) If that pattern holds true, then the nation would need to shift Dem by 8%, which is wildly improbable, since an aggregate of general election polls (Biden vs Trump/DeSantis) has the parties about even, with a slight R lead. The political climate is not what it was in 2008.
So, if it includes Biden, I give Alaska a VERY low chance of swinging. If it doesn't include Biden or Trump, I'd say your figure is more accurate, though I'm tempted to say it is still an overestimate.
Also, Mary Peltola only won because two Republicans split the vote. The sum of their votes was 49%, while she received 48.8%.
@ShadowyZephyr Biden and Trump are both unpopular, but Trump is less popular, with 55% unfavorablity as opposed to Biden's 53%, so I figure that neither candidate will be much more unpopular than the other if we get a rematch (which seems likely).
Races in which a candidate runs for re-election don't seem to swing nearly as much. (2004, 2012, 2020).
That's true - my quick analysis excluded 2000 and 2008 partially for that reason.
If that pattern holds true
The pattern may or may not hold true; compared to the country, Alaska swung 2 points left in 2020, 1 point left in 2016, 11 points left in 2012, 6 points right in 2008, 8 points left in 2004, and 5 points right in 2000. It's not all that consistent. Trends can be weird from year-to-year, e.g. Alaska in the 2000-2012 elections kept having big trends in either direction, or Iowa trending right by less than 1 point in 2012 and 13 points in 2016, or West Virginia trending right by 13 points in 2016 and then trending left by less than 1 point in 2020, etc.
Also, Mary Peltola only won because two Republicans split the vote. The sum of their votes was 49%, while she received 48.8%.
Alaska uses instant runoff elections to select its Representative, and she won 55% of the vote in the instant runoff election between her and Palin - i.e. 55% of voters preferred her to Palin. She would have won a one-on-one election with Palin, who would have been the nominee in the previous system. Potentially with Begich, too - e.g. a month before the election, Peltola's net approval was +26%, compared to Begich at +1% and Palin at -35%.
(Also, ranked choice voting produces more democratic results with a lowercase D than our current plurality system does, but that's a whole other argument I don't want to get into.)
@ShadowyZephyr Ah, my bad on that. I looked up the Mary Peltola election on Wikipedia and what I found must have been the first round voting. (Approval voting > RCV anyways imo)
Trump's been consistently more unpopular in the polls than he actually ended up being. My belief that Alaska won't swing is mostly informed by how I think the political climate is in America now.
Also, Trump's unpopularity only rose because of his indictment, which I think won't affect his core voterbase.
From my perspective, Trump is a very "like him or hate him" candidate, so his core voters are unlikely to abandon him, where leftists voted for Biden to avoid electing Trump, not because they like him, so they'll be more likely to abandon him. It's mostly anecdotal evidence, but I have seen that trend with pretty much everyone I know.
I didn't realize Alaska swung so much left in 2012, though, so I concede that a Dem win in Alaska is more likely than I initially thought.
@ShadowyZephyr Yeah, I prefer approval voting too - I think it ends up being a way better way of selecting someone who's actually going to be decent at governing, since approval ratings are usually very closely tied to candidate quality.
It varies from state to state; IIRC, 2020 polls underestimated Trump in some states (e.g. North Carolina) while overestimating him in others (e.g. Colorado). 55% is a bit higher than usual for sure, but historically Trump's disapproval tends to be at or above 53%.
It could happen, but it's also been my experience that leftists (and a lot of center-left people and centrists, too) really, REALLY hate Trump. I'm a leftist, and I'll vote for Biden in 2024 if it means decreasing the chance of another term of Trump - the same is true of my partner. Hell, I'd vote for basically anyone even vaguely left of center if it was to stop Trump from winning. I didn't like voting for Clinton in 2016, but I did it anyways.
Alaska is a weird, weird state politically, since no other state in the country is really anything like it demographically or geographically. The degree to which I'm uncertain about Alaskan politics is way higher than it is for any other state, so my gut instinct is always "there's a 10-20% chance of something really, really weird happening this election cycle."
Biden is not left of center anymore, if he ever was to begin with. From my perspective, I'd just vote third party because I as an individual won't change the result of my state.
I think maybe a 10% chance of a dem win in alaska is more realistic, then. FiveThirtyEight predicted a 15% chance of a dem win in alaska, and that was when dems were polling ~4% better than they are now.
@ShadowyZephyr Fair point, LMAO. I guess left-of-center by America's skewed standards. I've also only ever voted in "safe states," and the main reason I didn't vote Green or Socialist either time was just to, like...send a message, I guess. Decrease Trump's margins as much as possible in the already sapphire-blue cities I lived in? It's kinda pointless, but unfortunately voting for a third party feels pointless too. At least I'm a small part of the reason Colorado swung left by 8 percentage points in 2020...
Fair enough, yeah! The main reason I'm at 15% right now is just that the election is so far out. If the national numbers look like they do now in May of next year, I'll definitely be selling down to like 10%.