Betting odds are like, the most accurate way to determine outcome? Even for sports.
If polls were accurate, anyone with a bit of risk tolerance could make bank betting against Trump. But people doing stats in swing states, campaign finances, etc seem to be in alignment (again, if they weren’t in alignment, there would be money to be made)
From my reading, these betting markets have recently tracked as well as 538 long term, so that’s a decent outcome. I wonder if they just swing to vendors like 538 to look for “an inside track” to bet against overall, but it’s a small system so it’s not clear.
The issues they have are being largely illegal or grey market in the US, which limits who will interact with them. There very well could be experts who could “make money” that won’t do so because of their current status.
They also appear to be very erratic in the short term, the same way a lot of sports betting and stock gambling is. Single events can send the numbers into tailspins. We’ve had some pretty intense recent events, so I’m not exactly confident in their choices.
I’d say lastly that they are heavily right wing or right wing adjacent spaces. Some people are just there to make money, but others are tossing it away on ideology, especially in these divisive times. Markets aren’t always rational actors, they just tend to be in aggregate.
I think I asked you this last time you brought this up and you didn’t reply. What were Biden’s odds of winning in 2020 on these same websites at 3 months out?
Obviously gamblers don’t decide elections but they do a pretty good job predicting election outcomes.
I don’t think I’m the same guy you had that discussion with (or if I am, I have no memory of it). Anyway, Polymarket (which I trust more) appears to have no history for resolved markets (as far as I can tell) and PredictIt (which is currently more optimistic about Harris than Polymarket) only shows 9/20/20 at the earliest (at which time it was Biden 57, Trump 46, Harris 4). I’m not sure what you’re getting at with this question.
Note: Yes, PredictIt didn’t add up to 100%. In theory that means you could have made money off of it risk-free but in practice it has a lot of overhead which prevents it from being an efficient market.
Okay, so you think gamblers are good at predicting elections?
Why do you think this when you can’t find data to back up their predictions? What kind of “prediction market” doesn’t keep data on past predictions? Seems like an easy way to hide failure to me.
Well, this guy does keep track and the record looks pretty good. The average of several different prediction markets on 8/5/20 had odds of Biden 59.1 Trump 37.8.
60% of what? No poll puts him at that level.
You the guy who thinks betting websites decide elections?
Betting odds are like, the most accurate way to determine outcome? Even for sports.
If polls were accurate, anyone with a bit of risk tolerance could make bank betting against Trump. But people doing stats in swing states, campaign finances, etc seem to be in alignment (again, if they weren’t in alignment, there would be money to be made)
From my reading, these betting markets have recently tracked as well as 538 long term, so that’s a decent outcome. I wonder if they just swing to vendors like 538 to look for “an inside track” to bet against overall, but it’s a small system so it’s not clear.
The issues they have are being largely illegal or grey market in the US, which limits who will interact with them. There very well could be experts who could “make money” that won’t do so because of their current status.
They also appear to be very erratic in the short term, the same way a lot of sports betting and stock gambling is. Single events can send the numbers into tailspins. We’ve had some pretty intense recent events, so I’m not exactly confident in their choices.
I’d say lastly that they are heavily right wing or right wing adjacent spaces. Some people are just there to make money, but others are tossing it away on ideology, especially in these divisive times. Markets aren’t always rational actors, they just tend to be in aggregate.
I do trust people who are willing to put their money where their mouth is more than I trust people who aren’t.
So yes, you think gamblers decide elections.
I think I asked you this last time you brought this up and you didn’t reply. What were Biden’s odds of winning in 2020 on these same websites at 3 months out?
Obviously gamblers don’t decide elections but they do a pretty good job predicting election outcomes.
I don’t think I’m the same guy you had that discussion with (or if I am, I have no memory of it). Anyway, Polymarket (which I trust more) appears to have no history for resolved markets (as far as I can tell) and PredictIt (which is currently more optimistic about Harris than Polymarket) only shows 9/20/20 at the earliest (at which time it was Biden 57, Trump 46, Harris 4). I’m not sure what you’re getting at with this question.
Note: Yes, PredictIt didn’t add up to 100%. In theory that means you could have made money off of it risk-free but in practice it has a lot of overhead which prevents it from being an efficient market.
Okay, so you think gamblers are good at predicting elections?
Why do you think this when you can’t find data to back up their predictions? What kind of “prediction market” doesn’t keep data on past predictions? Seems like an easy way to hide failure to me.
Well, this guy does keep track and the record looks pretty good. The average of several different prediction markets on 8/5/20 had odds of Biden 59.1 Trump 37.8.
Interesting. Good to see data enter the conversation finally. It looks like on average, betting markets are about as good as Nate silver.
They also appear to have issues with legality in the US, which limits their effectiveness as a market, and they tend to be very “vibes” based in the short term.
I think that’s a good assessment of the current state of affairs.