*That’s what people who’s entire profession is establishing likelihood of outcomes think.
Oddsmakers are often wrong, but over the long term, they’re more often right, it’s the entire basis of how they make money.
Polls are just polls. Oddsmakers literally are putting thier money where their mouth is. If you’re confident they’re wrong, take the bet. They WANT you to.
Edit: after reading the great responses, I think I’m sorely underestimating the volume of bets and how keeping both sides betting against eachother in this case is the strongest factor in the current odds.
If people predominantly bet for one side, the odds have to lower so the house still wins in either case. Basically, there has to be enough taken in on all other options to cover the payout of any winning option.
If there’s no selection bias in betting skill level of the players, then the betting odds should roughly reflect the actual probability. But if one side’s bet has no basis in reality, then the odds can get very skewed.
And in those cases, it’s not the oddsmakers that are wrong, it’s the betters. The house always wins.
In a horse race, punters tend to spread bets across horses with no bias or favouritism - they place the bet because they want to make money, not because they are invested in the outcome.
In a political race, people bet for one team because they are ideologically aligned and want to show support.
If Republicans are much more likely than Democrats to gamble and place bets on their candidate, this creates market pressure and the odds for a Republican win will increase (I.e. get more likely) as a result of that.
As if polls aren’t designed and administered by people who know what they’re doing, and it doesn’t make them right. There never will be a way to accurately predict the future so why try? Vote.
Polls are surveys. They answer the question “what do people say they’re going to do?”
Extrapolating results strictly from polling comes with a pretty large caveat: it assumes people are going to do what they say they’re going to do.
Another large caveat is that there are issues with representation, as the group of people who will actually answer their phone is a pretty serious demographic skew.
Oddsmakers use polling results as a COMPONENT of their determination, but they’re free to aggregate against whatever else they see as relevant, things like history. Or to weigh which states polling is most effective. Or how which day of the week affects voter turnout on a per district level. Or how projected road closures could affect transit and therefore turnout rates.
Anyways, I think the entire point of my initial comment was lost:
If you’re excited about the polling, awesome… BUT DON’T GET COMPLACENT AND STILL GO VOTE. Why? Because the people who actually get paid to project outcomes think it’ll be a Hillary re-run, and if you don’t want that, you can NOT get complacent. You can’t let your friends get complacent. Be the reason the oddsmakers lose money in November.
I think we’re on the same page about what matters. I just don’t really put any stock in any prediction after what happened in 2016.
And yes I have heard the explanations and how actually the polls weren’t as wrong as they seemed at first glance. I just see zero reason to put any stock in them. I’m voting either way and I’ll encourage others to.
I’d love it if everyone at least was skeptical of polls because that’s a stance you can’t go wrong with.
I think the availability of polling is a major reason why people don’t bother to vote. If they’re reasonably sure their person will win, they’re less likely to bother. Or, if it looks like it’s going to be a blowout and they don’t think one vote will matter, same thing.
Voting day should be a federal holiday, there should be a tax incentive to vote like in Australia, and polling reporting should be delayed by 1 month.
For transparency, I’m Canadian, but our situation is pretty much the same: Far right candidate projected to win next election, “meh, status quo centrist” in power now, people letting projections deter them from voting, primitive “electoral college” lite making your votes less effective based on where you live, etc etc…
We do trend towards “looking over the fence” politically. If Biden’s step-down gamble pays off, it’s possible we could get a similar thing to happen here which could be good. So, we’re all watching and hoping that we don’t end up neighbouring a country with a fascist leader…
I hope for yours and the world’s sake that it works out on both sides of the border. I fear the US has a lot of negative influence on the rest of the world. Like if there was no president trump, how would elections have gone around the world the past few years? It feels like we set trends sometimes and some of them are amazingly bad
*That’s what people who’s entire profession is establishing likelihood of outcomes think.
Oddsmakers are often wrong, but over the long term, they’re more often right, it’s the entire basis of how they make money.
Polls are just polls. Oddsmakers literally are putting thier money where their mouth is. If you’re confident they’re wrong, take the bet. They WANT you to.
Edit: after reading the great responses, I think I’m sorely underestimating the volume of bets and how keeping both sides betting against eachother in this case is the strongest factor in the current odds.
I don’t think that’s how that works.
If people predominantly bet for one side, the odds have to lower so the house still wins in either case. Basically, there has to be enough taken in on all other options to cover the payout of any winning option.
If there’s no selection bias in betting skill level of the players, then the betting odds should roughly reflect the actual probability. But if one side’s bet has no basis in reality, then the odds can get very skewed.
And in those cases, it’s not the oddsmakers that are wrong, it’s the betters. The house always wins.
Exactly.
In a horse race, punters tend to spread bets across horses with no bias or favouritism - they place the bet because they want to make money, not because they are invested in the outcome.
In a political race, people bet for one team because they are ideologically aligned and want to show support.
If Republicans are much more likely than Democrats to gamble and place bets on their candidate, this creates market pressure and the odds for a Republican win will increase (I.e. get more likely) as a result of that.
How long have you been trading?
As if polls aren’t designed and administered by people who know what they’re doing, and it doesn’t make them right. There never will be a way to accurately predict the future so why try? Vote.
Polls are surveys. They answer the question “what do people say they’re going to do?”
Extrapolating results strictly from polling comes with a pretty large caveat: it assumes people are going to do what they say they’re going to do.
Another large caveat is that there are issues with representation, as the group of people who will actually answer their phone is a pretty serious demographic skew.
Oddsmakers use polling results as a COMPONENT of their determination, but they’re free to aggregate against whatever else they see as relevant, things like history. Or to weigh which states polling is most effective. Or how which day of the week affects voter turnout on a per district level. Or how projected road closures could affect transit and therefore turnout rates.
Anyways, I think the entire point of my initial comment was lost:
If you’re excited about the polling, awesome… BUT DON’T GET COMPLACENT AND STILL GO VOTE. Why? Because the people who actually get paid to project outcomes think it’ll be a Hillary re-run, and if you don’t want that, you can NOT get complacent. You can’t let your friends get complacent. Be the reason the oddsmakers lose money in November.
I think we’re on the same page about what matters. I just don’t really put any stock in any prediction after what happened in 2016.
And yes I have heard the explanations and how actually the polls weren’t as wrong as they seemed at first glance. I just see zero reason to put any stock in them. I’m voting either way and I’ll encourage others to.
I’d love it if everyone at least was skeptical of polls because that’s a stance you can’t go wrong with.
I think the availability of polling is a major reason why people don’t bother to vote. If they’re reasonably sure their person will win, they’re less likely to bother. Or, if it looks like it’s going to be a blowout and they don’t think one vote will matter, same thing.
Voting day should be a federal holiday, there should be a tax incentive to vote like in Australia, and polling reporting should be delayed by 1 month.
I agree with all of this completely! Maybe someday we can catch up to other countries…
For transparency, I’m Canadian, but our situation is pretty much the same: Far right candidate projected to win next election, “meh, status quo centrist” in power now, people letting projections deter them from voting, primitive “electoral college” lite making your votes less effective based on where you live, etc etc…
We do trend towards “looking over the fence” politically. If Biden’s step-down gamble pays off, it’s possible we could get a similar thing to happen here which could be good. So, we’re all watching and hoping that we don’t end up neighbouring a country with a fascist leader…
I hope for yours and the world’s sake that it works out on both sides of the border. I fear the US has a lot of negative influence on the rest of the world. Like if there was no president trump, how would elections have gone around the world the past few years? It feels like we set trends sometimes and some of them are amazingly bad
What were the odds Biden won in 2020? Did these same “expert oddsmakers” favor him or Trump?
Check the quality of your authority figures before you go all in on arguing they are right.