I'm open to people reporting that they lost > 1000 mana. I give a few days, and if I hear nothing, I resolve this as NO.

After the dangerous start, I thought this would be me, but was net -113 in the end.

@capybara 😬

At the moment Bohaskas anti-guess has 26/35 correct. The market before resolution has 24/35 and the Market before the very first resolution has only 15/35 plus three times 50%.

@nanob0nus
Bohaska: 28/37
Market's last guess: 26/37

Market's initial guess: 16/37

congrats @bohaska

Bohaska 30/39

Markets last: 27/39

Markets initial: 16/39

This but for 70% would have already resolved YES

Highlighting that I added an "are y'all any good at this" market

I assume that this is after everything has been resolved (so that if someone is temporarily a lot down but come back up again, that won't count). Correct if I'm wrong.

@Loppukilpailija Yeah. That's what I had in mind

😅

I’m currently down ~400 after two resolves.

Not sure this will be easy to check. But at least everyone should see their own total profit in this part of UI:

But if we're not sure, we can resolve this one to N/A

@roma I think at the moment the resolved questions are not included in those numbers, but in the end there will hopefully be a total profit ranking of all participants, like for the binary markets.

@nanob0nus Not sure. I've checked this one https://manifold.markets/Loppukilpailija/how-many-steps-are-needed-to-distin

It shows top traders like this:

But it doesn't show that I've lost 49 mana as in the above screenshot.

Although maybe these new unlinked ones show it.

@roma meh, only top 5 traders are shown. One can go to the profile of all traders manually and check their profits. But most likely someone who qualifies will bet this up and show proof for profit.

@nanob0nus Yeah, I'm fine just resolving this "NO" in case no one claims >1000 loss shortly after closing.

@Loppukilpailija did you recieve guesses for this?

@nanob0nus No one explicitly said that their submissions were about that - but the only person from which I got at most 100 predictions was Roma with their comment at https://manifold.markets/Loppukilpailija/which-of-these-random-bit-strings-a#5EyHmFJvqtdN78Ey6uVJ. I err on the side of counting those as submissions for this. Those are the only ones I'm aware of.

(Now would be a good time to say that "I did, in fact, send you submissions, but you haven't acknowledged them".)

@Loppukilpailija As it is now known that each of the guesses by Roma have at least 5 positions where they have been wrong, I resolved this market NO.

Do people have opinions on whether I should close (but not resolve) this market, and other markets that are about the time Just Before Resolving Anything?

@Loppukilpailija Oh, wait, I probably can't do that? Unlinked multi-choice doesn't support it I think. Gonna write a suggestion to Discord.

This one seems like a fun and good operationalization for "how confident will the market be as we reach the end?". Like, if you think that at the end the 40th market will e.g. be either 10% or 90%, you should bet this meta market to 90%

Does "max 100 predictions per person" mean that I can give 100 predictions at once and only one of them has to be correct?

@nanob0nus That is correct.

Can you give a formula for the average logarithmic score?

@nanob0nus log(p/(1-p)) when YES and log((1-p)/p) when NO, where p is the price at close? And in the end you check if the average of these 40 values is greater than 0?

@nanob0nus Not quite.

The logarithmic score is log(p) if the answer is YES and log(1 - p) if the answer is NO. (See e.g. Wikipedia for more.) And then I check whether the average score is better than log(0.5). (I have not specified the base of the logarithm since it does not affect result.)