225
this post was submitted on 18 Oct 2024
225 points (97.5% liked)
People Twitter
5218 readers
2033 users here now
People tweeting stuff. We allow tweets from anyone.
RULES:
- Mark NSFW content.
- No doxxing people.
- Must be a tweet or similar
- No bullying or international politcs
- Be excellent to each other.
founded 1 year ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
Uh yeah ofcourse they dont actually predict the future. They do however punish people for guessing wrong which creates a sort of natural selection of people that predict well.
When you bet money on a bad prediction, someone else will happily bet against you with the same amount and take all your money once they are proven right.
Then you have no money and you cant bet anymore. So pretty dumb strategy long term. Someone using multiple accounts is not an issue because the limitation is how much money the person behind those accounts has.
There are similar huge bets in both sides. It will balance itself out soon enough. But generally predicting elections is a dumb thing to try tbh.
I mean that would be at least partially more true if it were a sequence of bets and results but in this case it’s a single bet which will take forever to resolve so all of that analysis is useless. I’m sorry it just is.
What betting markets give is a sentiment of who people (who bet) think will win, which is something, but it’s different from predicting who will actually win. Humans are crap at predicting shit, especially humans that bet and especially humans that bet in political outcomes.
A huge portion of bets are very biased because of political “teams” and beliefs. If people suck at betting in sports events I would argue they suck even more at betting at political events where their identify and beliefs and even more at play.
Polling already gives you flawed data, betting markets even more so.
As i said...
According to a bunch of reputable studies, the average, of the predictions/guesses that lots of people independently make on the same question, will always be better than most sophisticated prediction systems. Thats the effect that prediction markets try to leverage, but for elections this is probably just not gonna work.
Also only being able to analyze the accuracy every 4 years is shit for a study. If the market was big enough to allow for doing the same thing for a bunch of local elections, it could be interesting.