This market resolves based on the Politico Poll of Polls tracker. It will resolve to the highest single average voting intention for Reform UK up to the date of the next general election; and not before. The market end date will not be extended unless, for some reason the date of the next general election is later than the planned latest date of the 15th August 2029.
In the unlikely scenario that this is not available at time of resolution, it will resolve based on the most similar available 'average of polls' resource, for example this page on the economist. If in this unlikely scenario, the highest intention is lower than the minimum of the range on this market, the market will resolve to the lowest end of the market's range.
Update 2026-01-15 (PST) (AI summary of creator comment): The 'above 31%' option requires Reform UK to reach at least 32% in the Poll of Polls tracker to resolve YES. A reading of exactly 31% would not be sufficient.
1,000People are also trading
Want to be clear here @traders that 'above 31%' means just that. If the market closed today it would not resolve YES, it needs to be at least 32% for that to happen.
@Simon74fe loool what is going on! i suppose the fault of the market is really that there's no option for below 31 but its sorta a paradox because they've already got 31
@LukeShadwell Did you set the numeric range from 31-100 when setting up the market? That might've avoided this problem but honestly numeric markets always seem to give kinda whack expected values (although not quite this whack...) so I don't think it matters too much, it's pretty clear to anyone what is going on
@LukeShadwell How odd! Judging by how manifold often is I thought you probably had set it right -- I'm guessing it's a bug then, or both of us don't know what we're doing 🙃
@LukeShadwell “Above 31%” is trading at 54% which means a 46% chance the final number is <=31%. Where exactly it falls is weighted by the higher options.
@Noit Yes, but if a range of possible answers is set when making the question why can the predicted value be outside that range? In my mind if Manifold calculates the value ans anything below 31 (as has happened here) it should be cap at 31 as the range set is 31-100?
@JayDH there’s two options at market creation, buckets and thresholds. If buckets had been chosen then the lowest bucket would have been 31%+, and it’d be impossible for a sub-31% answer to be returned. Thresholds sets a list of thresholds with 31% as the lowest one. Implicit in that is that the lowest threshold may not be met, so you get this counter-intuitive result.
That said it’s not a clear case for the buckets variant type either. With that, the creator can’t resolve one bucket at a time. If Reform hit 31% tomorrow then that threshold can be resolved tomorrow. If this was a buckets market, 31% couldn’t resolve until the next election or Reform hitting above 80% (god help us).
I’m not sure what the best answer is. Maybe creating thresholds with e.g. 10% as the lowest option and then resolving met %ages immediately?