Resolves to the candidate who gains the most in the prediction markets for the 2024 US Presidential Election, using Election Betting Odds' aggregated market probabilities at https://electionbettingodds.com/President2024.html#chart, during the day after the first debate between Trump and Harris for the 2024 Presidential Election.
Specifically, resolves to the candidate with the highest absolute percentage gain between the start of the debate (scheduled for September 10 9pm ET) and 24 hours later.
So, this question is basically about who the markets assess as outperforming expectations the most, which isn't the same as who performed best in an absolute sense because it's relative to prior expectations.
Details
The data points used for resolution will be the latest data point before the specified times shown on the graph https://electionbettingodds.com/President2024.html#chart
In case there's a tie (using the most precise displayed probabilities, currently percentages with one decimal point), resolves to 50%.
This market resolves based on the first such debate, regardless of reschedulings or cancellations. In case no such debate occurs, resolves N/A.
1,000
3.00@MaximLott That's not what this market measures: it's "beats expectations relative to what's priced in at Election Betting Odds", not "wins over more voters". (If Election Betting Odds is approximately calibrated, this market should be near 50% regardless of the likelihood of candidates winning the debate in the latter sense!)
If you want to point to a market on the latter topic, consider:
Expectations are already priced in.
So, this question is basically about who the markets assess as outperforming expectations the most, which isn't the same as who performed best in an absolute sense because it's relative to prior expectations.
Under some assumptions, you might expect that correct price of this market to be 50%. I don't think that's strictly true for a few reasons, but it definitely should be closer to 50% than the actual "who will win" (based on polls or whatever) predictions.
@jack Well I don't know about that. The mean should be that it doesn't change things, but the median doesn't have to be. The median could be shifted toward Harris if there is some tail that is very favorable to Trump, for instance.
@nathanwei yeah that's one of the "under some assumptions" I mentioned (that the distribution is normal or whatever). The other main assumption is EMH of course. Neither is valid, but I think they are somewhat close to valid here - it would be quite surprising if this market were at 70% like the win according to polls markets
@jack Trump is more of a known quantity than Harris though, so it makes sense that Harris' debate appearance could hurt her more than help her. We know what Trump will say. So if Kamala's left tail is fatter than her right one, the median could exceed the mean.
Neither is valid, but I think they are somewhat close to valid here - it would be quite surprising if this market were at 70% like the win according to polls markets
Agreed in this case, although I think it’s interesting that the symmetric distribution ~probably wasn’t true for the Biden vs Trump debate. E.g. it’s extremely hard for me to imagine a version of that debate which provided a similarly large swing towards Biden. And the conventional wisdom entering the debate seemed to be roughly “Biden has a small gain if he can dispel fears, with the potential for a large loss” (which if true, is exactly where this market need not be 50/50).
I don’t think that asymmetric framing is nearly as clear in this debate (alrhoufh if there’s a direction I’d say people expect the potential Kamala gains to be more modest than her potential losses in a screwup, which assuming EMH would imply Kamala >50% for a gain). And in practice EMH seems like a similarly iffy assumptions, especially given that this can resolve based on a tiny shift