Manifold Stats Page 2024 Megamarket
Mini
8
Ṁ1951
Jan 1
41%
Less than 1000 DAUs (7d avg) on any day
31%
will the "Cash stats" be displayed as 0 before the soft-launch, at any point before market close?
24%
D1 Retention goes below 50%
20%
M1 Retention goes below 40%
19%
2000+ signups in a day
10%
Daily trades passes 40,000
Resolved
YES
$20,000+ of mana sold in the last 30d, at some point
Resolved
YES
$30,000+ of mana sold in the last 30d, at some point
Resolved
YES
D1 Retention (7d avg) goes below 60%
Resolved
YES
A day with above 1000 signups
Resolved
YES
Less than 500 comments for a day

Options resolve according to https://manifold.markets/stats.

There's a bunch of things I, and maybe others, want to ask and try to predict about various of the Manifold statistics over 2024, so here's a market for that.

Get
Ṁ1,000
and
S1.00
Sort by:
Less than 1000 DAUs (7d avg) on any day

@Bayesian Also this statistic is looking scary for Manifold

@ScipioFabius yeah it's somewhat over

D1 Retention (7d avg) goes below 60%
bought Ṁ225 D1 Retention (7d avg... YES

@Bayesian This aswell:

@ScipioFabius thanks!

A day with above 1000 signups
bought Ṁ237 A day with above 100... YES

@Bayesian This should count?

Daily trades passes 40,000

@Bayesian looks like we got pretty close on election day, but just under, in which case very unlikely in 2024

$30,000+ of mana sold in the last 30d, at some point
bought Ṁ345 $30,000+ of mana sol... YES

@Bayesian these resolve YES, ~35k

I want to ask about whether this specific accounting discrepancy will be 'healed' in some way. You can add it to your market if you think it is worth exploration.

There is a new-ish section on the stats page called "Transactions From Manifold". It has one entry in a graph for "Cash stats":

It shows 50 "CASH" for 9 September 2024. It doesn't say what the CASH was for, but that is not the part that interests me. (I think it should say CASH_BONUS?)

The part that interests me is the source of this number 50.

The 50 comes from `txn_summary_stats`, which has these two relevant entries:

It looks like, at two separate day/times, an entry in this summary table was created that says 25 amount of CASH was sent from BANK to USER with type CASH_BONUS. But the "start_time" and "end_time" are the same! It might be an error, duplicating the records for the summary page.

If we view the actual txns table, we see this instead:

There is only one related transaction. There are no other transactions in txns that use the token "CASH" right now.

And if we check the indicated user's balance, it has a balance of zero "CASH":

Even though there is nothing in txns or any other part of the system that seems to reflect this.

So:

  1. Will the cash balance summary chart correctly reflect the contents of txns?

  2. Will there be an explanation for the reduction of 25 CASH from this user's account?

Nothing about this situation gives me any confidence that this website should be doing anything at all that even smells like "pretending to deal with real money".

@Eliza If you accidentally awarded 25 cash to a user that they should not have, why is there not a reverse transaction in txns? Why is their balance just zero?

If you are going to randomly duplicate or otherwise mess up the statistics page for the history graphs, why should users trust any other part of the stats page?

It is very important that these pages are actually correct and truthful.

It is absolutely critical that the flows of "cash" can be actually trusted. No should use a website where your "money" can just disappear on a whim.

My guess is that was a temporary bug, and then they made an exception-action to make the change appear reverted to all interested parties, instead of doing the more complicated thing of actually reversing it properly, because that would have had other disadvantages like delays and stuff. I added the following option:
> will the "Cash stats" be displayed as 0 before the soft-launch, at any point before market close?

I think slight stats inaccuracies are probably fine, but idrk. we could make a market about whether a loss of trust about the flow of "cash" leads to a large decrease in the usage of the platform by a large fraction of the platform in 2024 or 2025 or something, I'd bet NO on that. I agree it's somewhat concerning though

@Bayesian

> exception-action to make the change appear reverted to all interested parties

Isn't this just code word for "we will happily adjust your banking balance up or down on a whim"? THERE ARE NO CONTROLS HERE

Less than 500 comments for a day
bought Ṁ300 Less than 500 commen... YES

@Bayesian looks like this is YES