Topic: Merit Stat from theymos data (Read 1476 times)

No updates?

Great work!  Very informative, will not anyone promptly open the base of forum participants?

I have skills to do statistical analysis on merit system, but I don't have skills to get detailed datasets.
In addition, I believe that no one can verify or confirm that those data and analyses published by someone in the forum are correct one.
My suggestion is, Theymos (if possible) should give forum users datasets which contain more basic variables than the current ones.

This is just a great example of what kind of quality content everybody wants to see on BTT. Amazing job with this merits summary. I am looking forward to an update with even more insights on the merits. I like to know the exact figures of merit system because this gives me an overview of what happens and how this influences particular ranks. Great job OP.

Thank you for your information.forum are more and more members. The forum is growing. I hope the forum and admins offer better policies to help newbie as well as those who do not know about merit or smerit. It will create quality articles and quality members. The forum will be stronger in the crypto world. cheer!!

This post is really cool, excellent zentdex....
I agree with those give their smerit to you for this post, if I have it I will send to you. Your post really show a high quality post based on data and analyzed with statistical approach. I am interested with the decline of smerit sent as I recently do not give my smerit as I think that smerit is limited and nothing left anymore. This reason probably felt by most of forum community.

Great job Zentdex!
I believe it is not easy to present your really high quality post, I am certainly that this is a kind of high quality post. Decline of smerit distribution as you concerned, is that related with the smerit availability in each of members? If the availability is getting lack, that is most probably the reason why is the decline shown.

You created a stunning post. Your analysis is excellent, I really liked the graphics and tables that clearly explain the topic. Despite how many people are now discussing merit system, it was difficult for me to find a really interesting and useful posts. But your analysis really clarifies the situation and shows how effective all the innovations are. Thank you for such an informative report.

Great imformation! It made me feel more interested in earning merit, even if it was quite difficult for the newbie.

I think you are slightly confused about how Merit is transfered. When you merit someone you give them sMerit from your account, not your Merit, so your hard earn Merit does not decrease. Basically, for every 2 Merits you earn you get 1 sMerit added to your account. You give away sMerit and the user that receives it receives Merit.

Example:
- User A has 50 Merit (25 sMerit if all is still conserved) and User B has 0 Merit.
- User A then Gives 4 sMerit to User B (you merit someone, but you transfer from your sMerit balance).
- The final result will be that User A will still have 50 Merit (and 21 sMerit to give away). User B will have 4 Merit (and now 2 sMerit to award).

Excellent work you deserved a merit for sure but I can't give you any. I see that there are not many posts merited especially on few boards. Such a huge decrease in merits shows that users want to wait and not lose all their precious merits at the start because we all have learned very fast that this is not easy to earn one merit even if you write quality posts and do what you can to be supportive on the forum.

I had no idea how many Merits circulating in the forum. I think that soon, a very interesting calculation to make, will be how many users are with the pending ranking for not having reached the quota of merits. Ranking up from Full to Legendary is a very difficult task. It would be interesting to compare how many members became legendary year by year. In 2018 we can have a lower number than 2016/2017.

Thursday at 3 o'clock. I'll always try to post something at this time. : D

Thanks for your sharing. The article helped people like me to know a lot more about merit. I will try to post more quality

Great post. All drilled-down stats are a smashing way of exploring the data and trying to uncover flaws or the magic of the system.

I wondered myself how much sMerit was being given in a circular manner, that is subject A gives it to subject B and subject B gives it to subject A (not necessarily in the same order of magnitude). This is not necessarily a wrong thing to do, since I may be merited by someone, checkout his posts, and decide to give him some sMerit for one of his good posts. It could also be an indicator to farmed accounts meriting each other or pacts between users.

Regardless, my focus was just on trying to see how often it happens, and not going in to the detail of whether it is justified or not.

Using the most recently updated Raw Data source that I’ve found (timeframe 24/01/2018 through to 09/03/2018) , I decided to study that period as a whole, without baring other variables than the Meriter, the Merited, and the sMerit assigned within that period (i.e. no breakdown by dates/times nor ranks – by the way I cannot see the Rank in the raw data).

The initial dataset data for the given period shows:
Transferred sMerit in period:     97.171
Number of sMerit Transaction:  41.700

What I created was a table with the following information:
User_from: Meriter
User_to: Merited
NumTimesGiven: Number of sMerit Transactions from Meriter to Merited
SumMeritGiven: sMerit Transferred from Meriter to Merited
NumTimesReceived: Number of sMerit Transactions from Merited to Meriter
SumMeritReceived: sMerit Transferred from Merited to Meriter
Percent_given_over_received: Percentage of Given sMetir over received sMerit between Meriter and Merited


The above table is the starting point, holding all the pairs of sMerit transferred between Meriter and Merited, and vice-versa. The problem is that each user is included in the table twice potentially (as a Meriter and as a Merited subject). I therefore created a derived table that avoided the above issue, making each person figure only once in the table for a given Meriter/Merited couple.

In global, of the 91.171 sMerits given, 87.183 where inventoried as “given” and 9.988 as “received back”. That should not be taken verbatim since the process of keeping only one record for each pair of Meriter/Merited users is arbitrary (i.e. A giving sMerit to B is a record, but I could also have chosen B giving to A).

Where I really wanted to get to was finding pairs that had a similar ratio of sMerit given than received.

First I deleted all records that sent sMerit from subject A to B, but did not receive sMerit from subject B to A. The net amount of records now goes down to 2.976, with 22.880 sMerits involved (23, 55%).

Examples:

user_from   user_to   NumTimesGiven   sumMeritGiven   numTimesReceived   sumMeritReceived   Percent_given_over_received
729995   1115343   2   77   1   5   1540,00%
822485   990983   5   55   2   55   100,00%
511933   822485   2   55   2   14   392,86%
181084   213773   2   50   1   50   100,00%
112286   135065   1   50   2   50   100,00%
944905   1076869   4   50   3   50   100,00%
913200   914767   1   50   1   50   100,00%
316540   377879   1   50   1   41   121,95%
1001911   1064065   2   50   1   41   121,95%
988901   1033117   9   50   8   37   135,14%
1005975   1093441   2   50   3   30   166,67%
842041   1123153   4   50   2   28   178,57%
221867   298905   1   50   1   26   192,31%
336733   373994   1   50   5   23   217,39%
756570   1035345   1   50   1   18   277,78%
831989   1350445   2   50   1   11   454,55%
389331   935622   1   50   1   10   500,00%
915311   944905   3   50   2   10   500,00%


If we narrow down the query to those records where Percent_given_over_received is within the range of 80..120% (that is, sMerit given from A to B is roughly what B receives from A), the number of records (cases) goes down from 2.976 to 1.294, with cases such as:

user_from   user_to   NumTimesGiven   sumMeritGiven   numTimesReceived   sumMeritReceived   Percent_given_over_received
822485   990983   5   55   2   55   100,00%
181084   213773   2   50   1   50   100,00%
112286   135065   1   50   2   50   100,00%
944905   1076869   4   50   3   50   100,00%
913200   914767   1   50   1   50   100,00%
95437   860708   3   39   2   38   102,63%
1018920   1035345   1   36   1   31   116,13%
979216   982124   1   34   2   38   89,47%
1036967   1122909   2   34   4   29   117,24%
101872   976210   6   33   17   30   110,00%
771836   882756   3   31   4   31   100,00%
964133   1334470   4   29   6   30   96,67%
98986   153656   7   28   2   30   93,33%
1052553   1113026   1   27   1   26   103,85%
976210   1764044   22   24   5   20   120,00%
934304   939859   2   20   3   20   100,00%
81995   662293   1   20   20   20   100,00%
202760   598528   1   20   1   20   100,00%

Being the aggregate summarized as follows:

sMerit NumCases
55   1
50   4
39   1
36   1
34   2
33   1
31   1
29   1
28   1
27   1
24   1
20   6
19   1
17   1
16   1
15   2
14   1
13   1
12   2
11   5
10   21
9   4
8   13
7   13
6   24
5   44
4   30
3   30
2   162
1   918
Grand Total   1.294

So in summary (working on this last table where Percent_given_over_received is within the range of 80..120%, since I was aiming studying a similar sMerit transfer/receive amount -> these I call paired sMerits):

- The vast majority of paired circular sMerits are relative to 1 or 2 sMerits: 83,46%
- High paired circular sMerit transfer/receive Txs occur in 4,33% of circular assigned paired sMerits (10 sMerits or above).

- From a global perspective, 23,55% of sMerit asigned seems to be involved in circular sMerit asignment.

A lot more can be done with more detailed raw data and the precious element of time which I lack..

Wow! You really did a great job!

I have this question: If we remove 25 Jan and 26 Jan from the data, would Thursday and Friday still be the days when the most sMerits are sent?

I also want to say that merit sources are probably staff and legendary members, and that's why those members sent the most merits, not to mention that legendary members started with the most sMerits to giveaway.


We may be able to calculate the sMerits of any user, I don't know if it would be easy to do for a large number of accounts but I know how to do it:
Every user started with 'a' sMerits, received 'b' merits and spent 'c' sMerits, so the number of sMerits 'x' a user has is:
x = a + b/2 - c
I'm ranked as a member and I initially received 1 sMerit, legendary members received 200 sMerits - QuestionAuthority stated - but if I remember correctly Theymos said that not all users of a certain rank started with the same amount of sMerits but they actually started with an amount of sMerits based on the number of activity points gained in the last year, so if a user has gained 0 activity in the last year he didn't receive any sMerits.

We may need more research to know exactly how many sMerits every member started with, or we could ask Theymos , but I just thought of giving you my 2 cents.

@OP

Since the Ranks really matters now, not like in the past, I wouldn't be surprised myself if the number of members ranking up would gradually decrease. This has been one of the goal of the System that has been taken into effect immediately by far.

All in all, its great to see infographs such as this, though the raw data is open to public you definitely put an effort into this one.



Well, if they awarded "x" amount of merits to a number of members and have awarded number "x" back, then it would replenish the exact amount that has been distributed.

In other case, they could just apply as Merit Sources if they really awarded members that are by the way really deserved to have Merits.

Thank you. As the time passes the initial sMerits are given causing a "Big Bang" and now things are getting into cruise mode. All new merit comes from sources and Theymos will be able to regulate the issuance rate and validate the effects. My guess is that in 3 or 6 months the system will undergo its first review and some decisions will be made, even if the decision is just continuing.

The system has also had the side effect of exposing a few farms out there.

hello there, i've done this in the past where you can find all links for mertit givers https://bitcointalksearch.org/topic/m.31449020

Can you get stats to see how many people ranked up and who ranked up? This is the best way to find abuser

The system should periodically give smerits to people that are awarded ''x'' amount of merits each month, that would be an easy and fair solution, in my opinion.

Those are UNIX timestamps. You can easily convert them to human date in various apps.
Example: 1519904498 = Thursday, March 1, 2018 11:41:38 AM

I'm not zentdex but I'll help you anyway. I'll provide you links that'll make you more familiar about the Merit System.
Here are the links glad it'll help:

Merit & new rank requirements
3 Merit Stat After a Month of Merit
Where the Merit Pours?

So maybe the next time we meet you will be better. If you would really study the System and just read those significant information flouring in these section, by then you will not be confused anymore.

Also I wanted to point out that, you need to work on your communication skills mate, for you to make constructive posts and eventually rank up. I'll provide you a link(s) that'll make you more familiar about the forum itself: Beginners & Help
Especially this thread: Newbies - Read before posting

Thank you for your appreciation. Just stay tune, watch and Practice, soon you will create more than I could do. Keep Learning.


You can now see it here: https://bitcointalksearch.org/topic/where-the-merit-pours-3093768


Thank you for your appreciation.
I agree with you; I so many great and informative thread created by those rank, most of the time they nail lower rank questions. But there are exceptions...  That is what you get thru experience and wide reading.


Thank you for your appreciation.
I hope it clarifies some of untold information about merits. I'll be that one with the golden heart.  


That is a good idea, but it will need more time and more digging. I think displaying their usernames is not necessary for now.


Thank you. There could only be two reasons for that; It's either he don't like the post (but I am wondering why he even bother to comment?) or he don't like information about merits.


A simple thank you will do. Anyway, thank you for your appreciation


Thank you for your appreciation. Yes, there are so much work to be done, but with millions of members start helping together to explain merit, the work could be easy.

Awesome work! Congrats!

I hope forum admins use this data for upgrading and improving this merit system!

In my opinion there are much work to be done!

1) There are still people that don't know how this works.
2) People think that giving merit will decrease their merit count. (don't know difference between merit and sMerit)
3) Many users do not care to give sMerit

I won't be discussing merit system here! This is not the point of the topic!
Just a little message for forum admins.

Great work @zentdex

I am amazed that you think the OP does not deserve your highly treasured merit. Not even a recommendation from you? Wow! Such a well thought out topic and research like this should get well-meaning merits - at all time. Of the 29 merits the OP received, yours isn't even there.

This is one of the major problems of the Merit system. Most people giving this merit have their favorite posters and buddies. Whether they post constructively or not, the Merit givers find a way to merit the post. Now, think of how many merits the OP would've got were the poster a username like Theymos, etc. I am not saying the mentioned users aren't great posters but they seem to be famous here. That's my observation and I stand to be corrected.

This is very ludicrous. But then there are certainly people who have this mindset, no doubt.

This is something the admin never wanted with the merit post. He has mentioned that one should not give merits to a post just because he agrees to it. Also, giving merits to friends is wrong in my opinion. If you are giving them a benefit for being a friend and awarding merit to their posts over that of a new member, it is  partiality for the new member. The new members will then feel discouraged to see this and even those capable of contributing valuable posts here will leave. I am sure the admin never wanted to achieve this with the merit system.

First, can you tell the usernames of those users you are talking about, I would love to see their posts, especially who became full members. Based on you recent merits sent I see one such member, https://bitcointalksearch.org/user/asapjoshyy-1048815 , but he has lot of negative trust so doubtful.

I do not think I will be applying for the offer you made but I hope someone in need makes a good use of it.

Thank you for sharing this data this really helps me to understand more about merits. As I read this post I was enlightened that merit is very essential to each users now. My hopes are so high that merits will be given to deserving people by users who has golden harts.     

Excellent post, highly informative. I think the reason Hero members receive the most merits is likely because they are more well known in the community, they have formed friendships and create acquaintances here on the forum which leads to them being looked favorably upon in the posts. Also, hero members are bound to give out the most merit as nearly all of the merit sources are at least hero or above. 

Just wanted to say thank you for offering up merit rewards for tasks completed.  It is also nice to see that you have ranked up several members and jr members.  Also, to the statistician up above, great post.  If I had some sMerit to give you, I would as this was a really good post.

I'm getting more source merits to send, I've already ranked up 2 Jr members into members or 2 members into full members. currently I don't have any idea on how to post a task with merits as the reward. if I ask for something complicated, nobody would know how to complete the task, if it is too easy, they might get undeserved merits.

I have already made up my mind about giving merits to members, I would give them a task to complete and they would earn some merits. I had 3 sMerits last time I checked, now I have 7. if the thing that you have done here was useful to me, I could have rewarded you with some merits. my first attempt of distributing merits got somebody a negative trust.

I wanted to ask people to post on my thread and have me viewing their posts. due to the limited source merits given to me, I couldn't offer something interesting for people. if you are up for it, I could reward you with 7 merits if you could PM me the numbers of empty blocks from block #20,000 to block 500,000. if you think that you are up for the task, post here and reserve the spot, I don't want several members doing the same thing and expect to have the reward.

Zentdex Are you data scientist? Are you related to Sentdex youtube account https://www.youtube.com/user/sentdex somehow? I watch it a lot   I'm learning hard to get this career path, thinking about to get MSc after Coursera and Udemy courses. Great job, I wish I could to wrangle the data like that someday!

Thank you for your appreciation.
This is based on my opinion but as what I see both of your question is could be answered by YES; People are getting short of sMerits and trying to give less. However, I have no proof how much remaining sMerits are left in each account to tell they are giving less.


Thank you for your appreciation.
The count includes theymos; I’ll better manually check those numbers next time.

Thank you for that suggestion, I’m working on that. I’ll update you once finished.


Thank you for your appreciation.
Yeah, you’re right, I’ll do it next time. For the moment, I have another post which shows the number of member getting merits and their ratio. You can see it in this link: https://bitcointalksearch.org/topic/merit-ratio-3071785

They are already into that. They are continuously looking for members who could be a possible merit sources.
https://bitcointalksearch.org/topic/m.31327729
There are even list of applicants which you can found here:
https://bitcointalksearch.org/topic/unofficial-merit-source-application-thread-2823944
There is another list but I can’t find it at the moment.

Thank you for your appreciation, comments and suggestion.
I’ll better include those figures next time. By the way I sent a PM to you for the Excel version of the file. Happy Analyzing!

Honestly, thanks for your very informative thread.
However, I would like to give you some suggestions to improve your thread next time. I really want to do statistical analysis but unfortunately I am not familiar with types of datasets from the forum.
Average figures don't tell us the truth in some cases, especially if the distributions of target variables are not normal. Even for cases which have normal distribution, we have to present both mean and +/- standard deviations.
In contrast, for those cases which have non-normal distributions, we have to use median and interquartile range.
I will highly appreciate anyone help to transfer data from the forum to excell, which I can use Stata software for analysis.

Thanks.

We really need to increase the number of merits in circulation. We should not create a situation where members do not have enough merits to send to a deserving person. I have seen merit giveaway threads where the member starting the thread had to lock it as his Smerits got spend on initial applicants. In order to increase the merits in circulation- two things can be done. First is increasing the number of merit sources. Admin added some but they are not sufficient. Another option is to increase the number of merits given to each merit source per month.

Thanks for this analysis. Good thing I went here in Meta and I found this thread. It has been a question for me if the merit system is effective, given justice or members who give merits just give it to those they like.

If there is a gradual decrease in the rate if giving sMerits, I hope the Admins, Staffs, Legendary and Hero Member create a thread to give their justifications with such actions. I know they have their own reasons why they stopped giving sMerits. Were they fed up in giving merits? Were they exhausted? Can't they find quality posts now? Or what? So members like me will be informed about their reasons and expectations for me to have merit/s.

Great info. so this means newbie and jr. send more merits than members.
I am just wondering if those are send merits who among received those merits.

We don't know in which sections are given the most smerits?

Great review!


How could there be 64 unique staff accounts, 15 global mods, and 3 admin? I think the numbers are a bit off. There are only about 30 staffs, 4 global mods, and 2 admins (well it's 3 if you count theymos_away).

Can say it about right, as I sent 2 merits being Member  But seriously, it is a great job, hopefully, it won't be left unnoticed by the community. I don't have merits left, unfortunately. But I don't like the trend, does it mean people getting short of sMerits? Or trying to give less?
I would like to see the spread of merits among the users. As in the real life, must be a few users get most the merits, like 80/20 rule. Or you could think GTFO and collect your own data 

I create these charts based on the post of theymos seeking for further analysis of data.



I do hope that these charts will give further information about the Merit System. Unlike before, I try to provide my analysis based on the following chart; It will great if intelligent forum members out there could further explain these charts and give review if ever I missed something.
Before anything else, the rank is based on February 27 rankings.

The first chart is all about sMerit send by a group of users from January 24 to March 1; it clearly shows that more people are excited to use their sMerit two days after its implementation. Legendary top at sending Merits this is due to the fact that they have more sMerit than the lower ranks. But there are days (8 days) which Hero Members gave most Merits than Legendary, this is synonymous with Full Member and Sr. Member; which there are 16 days Full Member sent most of sMerit than Sr. Member.

But as time goes by, these sMerits have been exhausted, or people gradually slow down in giving sMerits. Let’s take for example the Legendary members who give a total of 3,131 sMerits on Jan 25 down to a total of 233 on March 1 which is 97% decrease. Below is the table of sMerit sent decrease per rank:

It is notable that Hero Member decreases their rate of giving sMerit at almost 98%, while on the other hand Jr. Members only decline by nearly 94%. Maybe because of Jr. Members are not given sMerit to share with others. The results seem disturbing that is why I created this second graph which shows the Moving Average of sMerit since its inception.


The graph shows that the Average sMerit gradually decreases day by day. The highest moving average (6,796) peaked at Jan 25 while the lowest (2,394) is marked on March 1. From Jan 30 to Feb 13 the average decline between five to three percent (5-3%) per day. Moreover, on Feb 14 to March 1, the average decline per day shift between two or one percent (2-1%). These indicate that sending sMerit is pretty much stable enough or there are lacking sMerit to send. If the latter is right and the decrease remains at 2%, the theoretical moving average 1 sMerit per day will be reach on March 2, 2019.



After seeing the continuous decline, I go back to the data and check what could be the cause and possible solution to negate the phenomenon. Thus, I chart all the senders, and these is what I see.


The chart shows that most of the sMerit sent came from the Legendary, and the least came from the Administrator. If you are thinking that the Administrator is not sending enough, you should probably ask the question, how many Administrator are there? Thus, I counted the unique account for each rank or position and ended up this table below:


The table shows that the Administrators are working hard to send sMerit to the forum members. They are the most generous sMerit givers followed by the Staffs. However, I have found that people in a Member rank send sMerit less than for those who are in Jr. and Newbie rank. This could indicate that they are saving their sMerits. And I just though that, I am one of them. Lol. And by the way, just round up those average to represent actual sMerit.

 
The next question that runs in my mind is the time most of the sMerit were given.


The chart shows that most of the sMerits are given during 8 AM up to 12 PM forum time. It is 16 hours long of sMerit giving, which peaked during 3 PM and plunge during 4 AM. Do this mean forum member from Western give most of the credits compare to Southern? I think more demographic info of users could draw a conclusion in this matter.

I also tried to figure out which day of a week got the most number of sMerit

The chart shows that most of the sMerit are sent during Thursday and Friday; This increase is probably the effect of the two days sMerit sending frenzy (Jan 25 & Jan 26).

I hope this post helps, comments and suggestion are very more welcome.