note to self: no more trading on things that look too big to fail

@Marnix Probably a case where you're glad the system did not let you bet it to 99.9%?

@Eliza i think it wasn't the system, it was just my liquidity 😛

@mods @Manifold this was resolved early. there are still provisionals and now trump is under 20%

@nikki yikes...

@nikki the “% reporting” curse comes once again

@nikki How about we unresolve it pending the final result.

Standing-by with my new balance of -2.5k mana.

@snazzlePop

goddamit

@Eliza how does this work for the Sweepcash markets?

@dlin007

[re: re-resolving a sweepcash market]

@dlin007 I don't think I'm supposed to do or say anything 'official' about the Sweepcash side. Only the 'Admins' can do those. For the mana market, I will treat it just like I would do for any other mana market, and you as the creator should probably do the same.

@Eliza @Manifold actually initially resolved both markets so I'm happy to defer to them again. Mana market will mirror whatever happens in the sweepcash market

@snazzlePop @WilliamKiely is officially a wanted man

@dlin007 ya, Manifold seemed sliiiightly over-zealous in their urge to resolve markets rapidly, tbh. Given that there were about 100 sweepcash markets on the election, it's no surprise that 1 will be overturned. I think they should aim for like ~99.8% confidence rather than 99% confidence before resolving

@benshindel i dont really blame them in this instance though...the 'evidence' presented by @WilliamKiely seemed pretty airtight. I would have resolved it YES on that basis too on the next day as well. But yeah, maybe they need a higher confidence threshold for super close markets

@dlin007 I did indeed warn that "% reporting" is usually nonsense, but no one believed me

@benshindel Well I have learned something today, thank you :)

Rip my positive mana balance 🥲

@dlin007 Thanks for tagging me! If Manifold reverses the resolution, then I'll lose 139.55 Sweepstakes cash, which was 100% of my balance at the time of market resolution.

@dlin007 In my defense to "@WilliamKiely is officially a wanted man", I wasn't the one to resolve the market. I was just the user who pushed for market resolution. It was Manifold's flawed resolution procedures that led to the market's early resolution. 😂

@dlin007 re "@benshindel i dont really blame them in this instance though...the 'evidence' presented by @WilliamKiely seemed pretty airtight. I would have resolved it YES on that basis too on the next day as well. But yeah, maybe they need a higher confidence threshold for super close markets"

Actually, I don't think the evidence I presented was airtight. I linked to Philadelphia's website which clearly labeled the results as "unofficial."

And when I asked an admin to resolve the market on Discord they replied "Maybe I'm being blind but where does it say 100% of the vote has been counted?"

And I responded by saying "100% of the divisions reporting means 100% of the votes. Divisions report once they've finished counting their votes; they don't come back later and say we've forgotten a few votes. If this isn't sufficient for you to resolve, please let me know."

The admin said "ah ok gotcha, thanks for clarifying" and proceeded to resolve the market. They could have instead said 'Sorry, that isn't sufficient to resolve. I can't take your word for it that 100% of the divisions reporting means 100% of the votes are in; after all, you might be wrong. Plus the link you shared still says the results are unofficial; we need to wait until the results are official.'

Shortly after Manifold resolved this market, I checked the Philadelphia website again to see if the vote count had increased. (While I thought "100% of the divisions reporting means 100% of the votes" was probably still true, I obviously wasn't certain.) The vote total had changed, though was still above 20% for Trump. I said "Oops" to myself in acknowledgement that I had been mistaken about "100% of the divisions reporting means 100% of the votes", but didn't bother to comment to say that Manifold had resolved the market prematurely. In retrospect I wish I had, though I don't know if it would have helped with anything.

Mistakes happen, manifold will unresolve and it'll be fine. It's a pretty common scenario to look at a source and think it definitively provides the resolution, but for the source to be wrong or for someone to have misinterpreted it. That's a big reason why we have the ability to undo resolutions

@WilliamKiely I actually was on the fence bc I did see the unofficial but I googled for the results and it was still the first search result, a .gov website, and seemed very safe to resolve.

It was bound to happen at some point unless we want to wait weeks to resolve certain markets. Luckily the amount of sweepcash traded on this market was low.

@SirSalty While the Philadelphia government website definitely seems like an official source (whether it shows up as the first result on Google or not), I just disregarded the fact that it labeled the results it was sharing as unofficial. I wrote it off as a case of "the website editor is slow to update the part that says "unofficial" but will now that 100% of the results are in" rather than take it as a signal that the results were not finalized.