@PC I'm tempted to go all in on 23 NO, but I want mana for the rest of the month 😂

@parhizj If it isn't 23, then in less than 12 hours I'm sure you can sell your shares at a rate below 1%

@FlorisvanDoorn I'm not (yet) a big time market marker. I know it shouldn't be but selling NO shares by buying YES shares is confusing to me and I'm worried someone else will mess up the order in between times (I'll end up with bunch of worthless YES shares rather than selling my NO shares).

@parhizj nvm. forget me

Nvm

really need a day with engagement more than that of 18th Aug but less than 1351. 1350 would be fine/fun just for the laughs

Oops wrong market

@SemioticRivalry Nice hedge. Forgot to move this market. There is a possibility that it’a 19th and below 1350. I think you should be fine though

@PC Yeah, I figured there's a <1% chance of that

Let's find out...

@Joshua I can't wait for this to land on 1348 and absolutely fuck me

@parhizj bruh why

@nobody I think it's undervalued with the amount of uncertainty, even though I expect tomorrow it will increase.

Also, profit!

Payout Ṁ1,439 on YES, Ṁ140

Also, I had yesterday as the largest engagement for the first week in the 3 week window so I'm not counting it out.

(Aug 18):
2023-07-29T00:00:00-08:00 to 2023-08-05T00:00:00-08:00: % Superconductor sample Users Engaged (1 weeks): 23.0 %

(Aug 19)
2023-07-30T00:00:00-08:00 to 2023-08-06T00:00:00-08:00: % Superconductor sample Users Engaged (1 weeks): 22.4 %

The second week has more engagement for Aug 19 than Aug 18 so there is some uncertainty.

that cliff after 22 is so weirrrrrd what do y'all know I don't

@Stralor Maybe I'm doing the math wrong, double check this for me first:
Aug 22 covers engagement (daily active users) for this period [start, end)
2023-08-02T00:00:00-08:00 to 2023-08-23T00:00:00-08:00:
This excludes Aug. 1 which was the peak of DAU (1d)

@parhizj right but DAU is only spike, how many converted? and then how many continued to convert after the spike?

@Stralor Edit: Let's give an example:
Engagement is being active on 2 days for each week over 3 weeks. Call the list of users being active for 2 days on a particular week something like UA2W (Users active twice this week). Engagement is the count of users of the intersection of three consecutive weeks of UA2W (3 lists)

Let's look at the first UA2W:

UA2W for Aug. 1 - Aug 7
Aug. 1: 3465 DAU
For Aug. 2-7, lets say 20% (just an example) of the people on Aug. 1 will bet/trade/comment again during one of these days.
That gives an estimate of UA2W for Aug. 1-7: 693.

UA2W for Aug. 2 - Aug 8
Aug. 2: 2425 DAU

For Aug. 3-8, taking the 20% again....

That gives an estimate of UA2W for Aug. 2-8: 485

There is a slight problem with the logic behind this intuition and that it's possible that the estimate is poor if more users are twice active off the peak DAU for that week (this is more/less likely with how peaky it is). For instance...

Take UA2W for Aug. 2 - Aug. 8. You could take the 2425 DAU for Aug. 2 and assume that on Aug 3-8 (DAU 2156, 2203, 2007, 1919, 2089, 1999) more users were active twice in that period (average of 2062 DAU for those 6 days): assuming 34% of 2062 users are twice active in Aug 3-8 would put the UA2W at 701. In this example that is still a large difference from 20% to 34% to make this happen, so that is why I think it is significantly more likely that engagement for Aug. 22 will be lower than Aug. 21.

That 13% on Aug. 22 bothers me (think it should be quite a bit lower and not the max). Anyone else agree?

I thought that everyone could add answers in a market like this. Is that still the case or has this been changed recently?

@FlorisvanDoorn Oh, I guess I should've gone for "free response". Oops. Ok, then I'll just add all the answers myself.