Trump at 50% is quite far off IMO (I’d quote 25@35 for Trump). His approval ratings are terrible, he’s substantially behind in polls, the country is currently suffering from a pandemic that he substantially mismanaged, the economy is doing poorly, and there’s no particular reason to expect this to change by November.
What is going on here? Usually I assume the people buying Trump at 50% are either GOP partisans or people who don’t understand how betting works, but I seriously doubt Zvi falls into either category. I plan to be at least $8k net short Trump by November, FWIW (better places to bet money currently).
Boris at 80% is also surprisingly low; I’m pretty sure the base rate for prime ministers resigning early when their party has a majority is much lower than those odds imply, and there’s no particular reason for him to be unusually likely to leave. His approval ratings also aren’t unusually low (2 months ago I would have been worried about coronavirus, but he’s now immune).
There’s much more than a 10% chance of the race being undecided 24h after election night (especially if you think Trump has a 50% chance to win). For example, the results in the 2018 AZ senate race shifted roughly 3.5% from election night results to the final results, because AZ elections are largely conducted by mail. There’s probably a lot more vote by mail (VBM) this year (every potential swing state but texas has no excuse absentee voting by mail), and so many states are likely to count very slowly, just like AZ. If the election is within 2% in the tipping point state, then it seems pretty likely that the eventual winner will be under 95c on PredictIt 24h after the election. This is doubly true because many states are likely to be unprepared for the surge of absentee VBM ballots, and may take longer to process them.
I don’t really have anything to say about the others, but I spend a lot of time betting on politics and those seem pretty off.
(I’ll offer to bet with some people at much better odds than PredictIt on Trump losing, as a symbolic gesture against this really terrible conventional wisdom that he has a 50% chance. Not willing to bet on the others without doing more research)
Trump at 50% is quite far off IMO (I’d quote 25@35 for Trump).
That is, you would accept either of the following bets?
If Trump wins, you pay 25; if he loses, you gain 75
If Trump wins, you gain 35; if he loses, you pay 65
I would take the second one, my $35-70 if he wins against your $65-130 if he loses, if you can give evidence for your trustworthiness. (I can do the same. Nothing personal, just you don’t have a history here for me to go on. Lower stakes probably aren’t worth the hassle for me.) But fair warning, I would hedge it on betfair to win out either way.
Right, so I’m not exactly sure what I want to accomplish with my “symbolic statement” but betting with someone who immediately hedges it all out on betfair probably isn’t it :) I think I want to make it so that people who post what I consider poorly reasoned odds have money on the line if their reasoning is wrong, which isn’t accomplished by betting with a 3rd party who hedges their risk.
(also I didn’t read your post correctly the first time; the 2 bullet points would be correct if you replace “Trump” with “a candidate other than trump”)
Yep, i think that’s exactly why markets have Trump at 50% chance to win (that’s what I meant by “don’t understand how betting works). It baffles me that Zvi and Scott also have Trump at 50% chance to win though; it’s a dramatic and unjustified overcorrection.
Trump at 50% is quite far off IMO (I’d quote 25@35 for Trump). His approval ratings are terrible, he’s substantially behind in polls, the country is currently suffering from a pandemic that he substantially mismanaged, the economy is doing poorly, and there’s no particular reason to expect this to change by November.
What is going on here? Usually I assume the people buying Trump at 50% are either GOP partisans or people who don’t understand how betting works, but I seriously doubt Zvi falls into either category. I plan to be at least $8k net short Trump by November, FWIW (better places to bet money currently).
Boris at 80% is also surprisingly low; I’m pretty sure the base rate for prime ministers resigning early when their party has a majority is much lower than those odds imply, and there’s no particular reason for him to be unusually likely to leave. His approval ratings also aren’t unusually low (2 months ago I would have been worried about coronavirus, but he’s now immune).
There’s much more than a 10% chance of the race being undecided 24h after election night (especially if you think Trump has a 50% chance to win). For example, the results in the 2018 AZ senate race shifted roughly 3.5% from election night results to the final results, because AZ elections are largely conducted by mail. There’s probably a lot more vote by mail (VBM) this year (every potential swing state but texas has no excuse absentee voting by mail), and so many states are likely to count very slowly, just like AZ. If the election is within 2% in the tipping point state, then it seems pretty likely that the eventual winner will be under 95c on PredictIt 24h after the election. This is doubly true because many states are likely to be unprepared for the surge of absentee VBM ballots, and may take longer to process them.
I don’t really have anything to say about the others, but I spend a lot of time betting on politics and those seem pretty off.
(I’ll offer to bet with some people at much better odds than PredictIt on Trump losing, as a symbolic gesture against this really terrible conventional wisdom that he has a 50% chance. Not willing to bet on the others without doing more research)
That is, you would accept either of the following bets?
If Trump wins, you pay 25; if he loses, you gain 75
If Trump wins, you gain 35; if he loses, you pay 65
I would take the second one, my $35-70 if he wins against your $65-130 if he loses, if you can give evidence for your trustworthiness. (I can do the same. Nothing personal, just you don’t have a history here for me to go on. Lower stakes probably aren’t worth the hassle for me.) But fair warning, I would hedge it on betfair to win out either way.
Right, so I’m not exactly sure what I want to accomplish with my “symbolic statement” but betting with someone who immediately hedges it all out on betfair probably isn’t it :) I think I want to make it so that people who post what I consider poorly reasoned odds have money on the line if their reasoning is wrong, which isn’t accomplished by betting with a 3rd party who hedges their risk.
That’s fair :)
(also I didn’t read your post correctly the first time; the 2 bullet points would be correct if you replace “Trump” with “a candidate other than trump”)
Right, yes. I’ve never got the hang of translating things into bets. I spent several minutes checking and still got it wrong.
Before the election that Trump won a lot of people severely underrated Trumps chance of winning. Now, they seem to tried to correct for it.
Yep, i think that’s exactly why markets have Trump at 50% chance to win (that’s what I meant by “don’t understand how betting works). It baffles me that Zvi and Scott also have Trump at 50% chance to win though; it’s a dramatic and unjustified overcorrection.