Topic: Analysis of users after Merit kick-off - What the've been up to (Read 247 times)

alia received the majority of the merit before the negative trust. And additionally, post quality ≠ trustworthiness. Even TradeFortress could make a great thread. Most probably would be inclined not to give him merit, but it would be fine to do so.

---

Have to remember: you could gain a lot of merit but then still get negged afterward.

Looks like people give her red trust just for fun, as she is not participating in any signature campaign and obviously don't care. -1024 looks epic, can call it Gigatrust 

Nice analysis big contribution with a lot of hard work.  When first saw your very high merits I was surprised but really you deserve.  I don't have any to give otherwise definitely I would have sent.
Your analysis shows one DrMann having more than around 118 merits but almost all given by just one person.  I do not have any challenge and dispute about intelligence but my view is there are so many other intelligent persons whose contribution is very big, they should also have been given some share of one's merit stocks. 
One profile alia on the one hand is very heavily negative red trust -1024 but  with merit score 113 though the person is very new entrant to the forum  i.e. entered in January 2018.  If such person is so meritorious so why people do not trust or negative trust him/her?
Above are just the questions which came to my mind; but great and big salute to your work.

1. Introduction

I've taken a look at the evolution in the forum of users that created a Bitcointalk account during the first two weeks after the merit system was set in motion, to see how those users born into the new system have developed. I must stress that this analysis is limited to new users created between the 24/01/2018 and the 06/02/2018, being the grand total 64.991 users created in those two weeks.
This is therefore not to be extrapolated to all users, since only new users with a 2 month window ahead are treated.

Those users have had time to become Members in these 11 weeks, so I wanted to see what had really happened. Not everyone has a fast track objective to rank-up, so this exercise may be different if performed in a couple of week's time with the same set of users.

Two quick conclusions:

a) Not that many users are really that active and eager to rank-up (generated posts could be way higher if that were the case).

b) Merit has it's toll as expected on ranking (see "9. Rank, Activity and Awarded Merit").



2. User Accounts created by date



The first thing that stands out is the amount of user account that were created on the first day of rank kick-off. The amount is double an average day. Maybe due to bots creating accounts, or to users rushing to see if they got in before the merit system actually was set in motion (in was booted up at night)... as it that were to be a benefit ..


3. User Accounts created per hour



Nothing strange here: accounts are created mainly between 8 am and 8 pm (forum time)


4. Last active week



This is where it gets interesting. It turns out that only 13,01% of these users have been active in the past two weeks, and an additional 5,65% two weeks before that. That is to say, only 18,66% of the newly created accounts analysed actually login to the forum in the past 4 weeks (let alone do something on it).

There's even a 10,96% of these accounts that have never been active (I guess creating the account and exiting does not leave a last active timestamp). Again that is fairly odd and may be due to bots as TheQuin suggested answering one of my previous posts.


5. Posts per User



The first thing that stands out is that 75,13% of newly created users have 0 posts (I've omitted this from the chart to be able to graph the rest, but the table contains this data).
WTF? That is an outstanding figure. We thought that every user would post at least a few times in the last 11 weeks, but the immense majority are just voyeurs (gaining knowledge let's hope) ... or bot accounts...

Let's say that a post a day is what is needed to contribute to activity and therefore rank (merit aside). It turns out that 2,19% of the new accounts had more than 51 posts in these past 11 weeks. Not that much effort overall in ranking up so if we complain now of the amount of non-contributive posts created on a daily basis, imagine if the whole base of newly created accounts really did go wild on trying to rank up with daily activity!


6. Activity per User



Not much here once commented the above, since activity is correlated to posts. I classified activity for a table that is shown further down as high (1,47%), mid (2%) , low (21,40%) and none (75,14%). We could say that, from a ranking-up's perspective, only around 1,47% (high) are really involved and on route to rank up as fast as possible (merit aside).

P.D. There seems to be a small glitch that allows users with 0 activity to have >0 posts in their count (that's why there's a 5 user difference between the 0 activity and 0 posts groups).



7. Rank



Not surprisingly having seen the previous charts, there are not that many users that have ranked up to either Junior Member or Member: 13,72%. This doesn´t mean that the won't somewhere down the line.
For now we can at least say that only 0,27% (if we exclude the brand new from the ratio) made it to member on the fast track (being active often enough in the window of time analysed and gaining the 10 sMerits required at least).


8. Merit



Merit is kind of spread-out across ranks, but average per user is low. Of course is posts are low as seen before (phew), a large aggregate of merit is less likely to be gained.


9. Rank, Activity and Awarded Merit



note: horizontal->rank and categorised activity. vertical->merit group

Rank is achieved through activity and Merit, therefore they should be looked at together to comprehend the proper picture.

As it turns out, there are:
* 6     Jr members with high activity (more than 60), but need less than or equal to 4 merits to rank-up. -> These would have ranked-up with old system
* 112 Jr members with high activity (more than 60), but need more than or equal to 5 merits to rank-up. -> These would have ranked-up with old system
* 782 Jr members with high activity (more than 60), with 0 Merit -> These would have ranked-up with old system

There are four exceptional users with above 50 Merits:

https://bitcointalksearch.org/user/alia-1764044   alia                          Jr. Member
https://bitcointalksearch.org/user/esmanthra-1764764   esmanthra                Member
https://bitcointalksearch.org/user/battlestarcoin-1792612   Battlestar_Coin         Copper Member
https://bitcointalksearch.org/user/drmann-1806854   Dr.Mann                   Jr. Member

One of the profiles above is rather blurry with bags of negative trust...