https://manifold.markets/stats
An engaged user is a user who has traded in, commented on, or created a question on at least 2 out of 7 days in each of the past 3 weeks.
1,000🏅 Top traders
| # | Name | Total profit |
|---|---|---|
| 1 | Ṁ16,678 | |
| 2 | Ṁ9,281 | |
| 3 | Ṁ3,674 | |
| 4 | Ṁ2,203 | |
| 5 | Ṁ942 |
suggested by@nickten
In case you weren't sick of these markets yet. Given the spread between the 1300 and 1400 market, here is a market for 1350. Let's see whose models is most precise ;)
I tried to rule out (by plausibility) engagement in users from signups compared to the base:
https://github.com/JRPdata/wastebook/blob/main/manifold-engagement
Even for 1500 engagement I couldn't rule it out.
The limited sampling I did based on bets (separately from the above) didn't rule it out either: showed ~6 out of 27 people for n=100, bet at least (twice) 7 days apart starting from the superconducting boom period).
Based on this sparse data, I won't bet until at least (Edit:) August 8-9 when the retention for NW1 comes out, reflecting some information from the start of the boom.
For anyone relying on new user retention data for NW1 without reading the code for how it's computed, you probably don't know what it is doing (I guessed wrongly until I read the code):
My summary of what NW1 does:
For a day #,
Gets new users created between 7 days to 14 days prior to a day #
Gets active users from day # to 7 days prior to day #
Finds the active users that were retained (belonging) in this new user set.
Gives the % between the active retained to the new users.
So, NW1 is about % retention of users (active) last week from users created between 1-2 weeks ago.
@parhizj New user retention (NW1) For Aug 8 is > 30%. I take this is as a good sign and predict high engagement in a couple of weeks.
@PC Still I'm suspicious of the many YES bets. Surely not that many people don't understand its 2 days of activity in EACH of 3 weeks (an AND condition so they have to stick around for 3 weeks). You did right?