Topic: 100 Days of Merit (Read 1792 times)

Please stop call someone with prefix like Mr., Sir, gentleman, etc. It demonstrates that you tend to over-respecte someone you don't know. In addition, this sort of behavior is very common in beggars. After the launch of merit system, there are lots of merit beggars negatively fell into such stupid approach.
I mostly ignore threads which contains those words.

Actually the quality of a post is not the only factor I consider. A post may be good, but if I think the merit is going into the bounty spamming community, I'll probably pass on awarding merit. In this case, I thought the poster was joining the forum with an apparent attitude that I considered to be beneficial. I wanted to encourage him to continue. Hopefully his posting will help to drown the total rubbish produced by most new members.

In my own point of view, some smerits are not distributed properly in the forum because of smerit trades happening secretly. Some smerit are given to same person that received 50 merits. That is why some good quality, constructive, creative, informative poster won't have a chance earning merits even though they really contribute to the forum. Not all new accounts are dummy some are real and new person who join and trying to contribute in any way.

You,  Mr Cash, just proved that no matter on quality of posts giving of Merit is subjective decision, someone give Merit for informative posts and some just for support, just like in this current case.

Looks like you know your stuff, I'm just started to learn and didn't try anything special rather than importing data in csv form  Problem that theymos didn't offer data in such an easy format.

I thought once that zentdex is the owner of "sentdex" channel on youtube, but he didn't confirm that. But if I see a tutorial on how to get data from simple machines forum on that channel, I wouldn't be too surprised.

Of course, it is legally to publish non-constructive threads, but it is always good to say thank you to someone who help you to show your politeness.
Congratulations, PirateHunter. By now, you had 2 more merits and 1 more postcount.  

Can I say thanks? Or is it illegal?   

I gave you a couple of merits to encourage you to experiment with this. It would be interesting to see a new members perspective.

I would love to know How did you get the Data?

Did you use web scraping tools like (BeautifulSoup) in Python or is it available in the Forum. I am new here, I am AI Engineer, I can get some awesome insights on the forum with this kind of data. Please guide me in the right direction.

Really good work with Data Visualization. What did you use? (Python + Plotly or Matplitlib or R or something totally new.)

The merit system was meant to focus on the quality of posts, or the number of merits!

If gaining of merit gone well to me in the beginning now everything seems frozen. Probably the thing is that sMerit holders spent all of their sMerits in the most beginning. I was reading that some guy sent 70 sMerits to one post just to check out how it works on the first day of new Mery system, or maybe it simply was his second profile here How to earn Merit is the point that became very actual nowadays. As I see only threads that suggest to make the system harder get the most number of merits, such us https://bitcointalksearch.org/topic/m.32278787 and all those crying for merit posts stay death.

Hello zentdex,
Thank you very much for great topic!

By the way, I have small suggestion for you, which I think you should apply for your next topics/ threads.
In some parts of the topic, you attached oversize images (I quoted one of them as an example), which may lead to annoying feelings from readers.
In addition, resizing them to be a little bit smaller might be a better option.

You can use softwares to reduce image sizes, but in the forum, you can do it directly.
It's very simple, and you can visit my topic to read more about those tips.

Guideline on posting images (with size adjustments), hyperlinks

I adjust a bit to show your resized image, like this one:
Personally, this one is large enough for readers.

Another fruitful analysis. Thank you for providing us those details, there are user looking for more clarification and current state of the merit system. I never thought that merit giving is continuously decreasing. One thing that I notice, higher rank mostly prefer to give merit to those who are in the lower rank. BTW. I am currently in a Member rank providing help to lower and higher ranks by rewarding them a merit once they have answered a Bitcoin related Cryptographic Puzzle. By this it also add another knowledge not just to the winner but to the thread readers. I will be very glad if you sponsor even one puzzle. You can visit the thread here at https://bitcointalksearch.org/topic/m.32048143
Thank you.

@zentdex Thank you for the analysis, I enjoyed reading through your post although I was finding it difficult to understand the Graph but I enjoyed every minutes I spend reading. Keep it up. Thanks for your contribution to the community

I agree to disagree with you   Nope, it's not just because they can't find any posts worthy of their merits but also because, most times those useful post are on threads they're not interested in. Therefore they won't be merited for their contribution to the community.
Examples: while surfing through threads for useful information, i saw a warning about a new type of scam. I took precautions and it helped. (It was when the Giveaway scams on popular Twitter accounts were still new.) But to my surprise no merit was given to the posted and the post was 23days old with 60+ comments.

For statistical presentation,
With quantitative variables, in the case of your topic, merit points; those sort of variables should be presented in the following style:
Average +/- standard deviation; and Median (Interquartile range).
Average and Median are not enough, and might lead to misinterpretation on real distribution of merit variable.
Median and Interquartile range are essential statistics for non-normal distribution variables.
You choose right statistics to desribe your variables, zentdex.

For your interest, I also give you link to helpful statistical guides from laerd:
Measures of central tendency

Mean, median, and mode should be used correctly with related supplemental statistics for central descriptions.

Hope it helps, zentdex.

Thank you very much, zentdex, for doing great analysis.
In addition, you likely spent lots of time to retrieve data for your analysis based on raw data Theymos gave.
Hope that you will release more meaningful threads in the future for bitcointalk forum's community.

What a thread! The stats were easy to interpret and understand, I wonder why is it that giving sMerits decreased as time goes by in this forum. Maybe because merit givers cannot find any posts that are worthy of their merits..  Is it true? 

Merit Average
Based on the table below it clearly shows that Newbie and Jr. Member have an average of 3 merits received in the span of 100 days.
It could be assumed that they could easily rank up to a member rank.


Well, i guess i need to wait more that 300 days to rank up. I hope not though.

I just like the way you describe the whole things.
all the graph are so informative and very helpful for us.  actually your presentation really deserve more view and merit.
I hope we will get something more like this.
thanks very much. 

all graph seems to be very informative, it shows how this merits affect each member. most of the  graph dominated by a higher ranks in giving merits to everyone. you can also determine that full member and above are the most active in giving merits to a person that they think they deserved to have merits. it also shows that merit system become more effective to avoid shitposter to rank up their account.

The presentation is very informative, although OP did not conclude but allow us to form our opinion on the presentation. There are observations I made: The merit score has in more way than one, limited creation of multiple accounts. It is steadily instilling the culture of quality and constructive posts. I am sure the OP had gone into research of this post for the quality of it and for the improvement of the forum, which is a value of the merit system.

Had the merit system not been introduced, by now the stats would have been over 20,000 new accounts created per day.

The forum should keep this merit system but develop a strategy to checkmate the cheats.

Thank you. It continuously decreasing indeed.


Yes, information here could be theoretically correct.


Just a clarification, I am not telling that there are something wrong about it. It could maybe go according to their plan.



Good observation, I have draft to do your observation and I see some probability.
I better study and create something like that.


Yes I think we need more merit source and more merit giver...


I forgot to include those; Actually, I have done it before in Where the Merit Pours?


I agree with that notion.


Maybe, no one knows what are the other plans of theymos in this merit system


I try to check the figure but, yes those are right. However, merits on the first three days were excluded.


Yes, it could be a factor.

Good work zentdex and keep it up, it was a very informative post. So now we can say that introduction of merit system has really helped to control the creation of alt accounts. It would be really interesting to see if we get the data of those users who are frequently exchanging the merits between the fixed users. 

The system really did work out fabulously. Also that "black market" will definitely be a problem and there is no way it will be tolerated. It would be completely unfair to buy your way up while others have to contribute their time and knowledge to earn their fair share of merit.

The Merit system was a very good decision. Otherwise, spamming started a lot. There are currently too, but very few. Moreover, the account sellers are  in great danger. They are not able to raise account rank again. I have seen some accounts in some bounties, which were created in 2010 but are being used for 2 months. I certainly welcome the launch of the system.

I really appreciate for your informative thread and graph. I have read also your other thread about merit explrination. Really merit system reduce spamming and increase HQ post. Members must be make quality post to get merit.

Not really, merit is decreasing but as you can see from this Table (hoping I will do the format in good way), higher ranks needs a lot more merit than Jr

_____________	________________	______________
Rank	Activity min	Merit to earn from past rank
Brand new	0	0
Newbie	1	0
Jr Member	30	0
Member	60	10
Full Member	120	90
Sr. Member	240	150
Hero Member	480	250
Legendary	775-1030	500

Do you still think it's difficult when one has to earn the same merit 50 times?
Certainly a hero could be helped in the sense that he is more informed about this world but the difference is abysmal expecially now that the airdrop is almost gone.

Thank you author for the informative topic. I think the hardest thing is to raise the rank of Junior member. Many do not give a low rank measure because they think it is another person with multiple accounts.

\\\\\\\\Merits in the long run well be depleted, so it become more harder to get the rank that you want even if you try hard to get that. Ergo, all well gonna stuck in the position were standing. Only few merit sources against thousands of post in the forum trying to get 1. Well Good Luck to us!

Maybe this what theymos planning all along, multiple accounts well never be gone, but atleast this may slow them down. I can't wait to see the system in it's anniversary special.

Theymos actually deserved those merit points due to his long-lasting contributions to the forum, despite the fact that he doesn't need those merits.
The forum members have their own rights to send their sMerits to Theymos if they want.

This is actually a great thing that you have reacted so fast to the number of new accounts increasing. Could that be due to bounties as well as they are getting more and more popular and this forum is the main source of it all?

From that perspective, it's a good thing that it's harder to rank up as it will filter the members who are hard working from the competition that spams and doesn't add any value at all. I really believe bounties are the future of work, giving opportunities to people from all over the world to make extra income.

Perhaps it would be interesting to see which sections get the most attention when it comes to giving and receiving merit?

Also, have you considered that a "black market" might emerge... people buying and selling merits OTC? It sounds silly but who knows...

Thanks a lot for the detailed stats

Interesting fact, I assume you're basing your stats on forum time so, it could be that Sunday US time is the best time for getting merit
I see you posted on a Sunday too 

From this table it looks like the more of a higher ranked member you are, the least important merit becomes

I would be curious how many merit transactions were created for high amounts of merit. Let's say over 20 merits/transaction. Could you provide something like this?

Another great data presentation.

This could be a basis where the member are paying more attention for healthy discussion. We can also see the topics that topped at merit for a certain section and also the poster.

Giving and receiving of merits are the most important data so far. I think Senior and Hero members should start working on some merit distribution to add more stats on members that really deserve a merit.

@zentdex

Thanks for all the charts and info, good work mate!

Regarding decrease of new members registering per day, the January peak match almost exactly crypto marketcap ATH and after that it starts going down along with the whole market, even before merit system implementation. I wonder what has a greater effect on registrations number - merit and account rank up difficulties or bitcoin and altcoins price.

Thank you. You are right, the number of member who give 50 merits in a single transaction decreases.


That is also what comes in my mind when I have see the values. But just like you said, it will be hard to determine which are alt accounts even considering IP. Thank you.


I use 1030, because It's really hard to pinpoint which is which.


I'm with you by that. There are estimated more than 1,000 accounts who received merits but did not give back enough to others.


Good suggestion, thank you for that. It could took me a longer time to do that and may require a greater computing power.


Because the dataset is not what other user expected to be. I could be more harder to understand. For example, I have a separate table for user id and name, another for merit, another for post, activity and rank, another one for dates, table for who send and who received among others. I then create a queries to see what I am looking.

• Average amount of merit given, per transaction?  - 2.16 with deviation of 3.97
• Average of merit given per transaction, by rank level - see table above
• Can we isolate transactions by forum board? Some boards are more likely to have these questionable transactions (e.g., alternative currencies, local, etc.) - It is possible, but what is the purpose of isolating those boards?
• We should isolate the received side of the transaction, an alt account is more likely to receive transactions that are arbitrary gifts than due to genuine quality of their posts - obviously just an unsupported hunch on my part -I have stat that dived those two. see II. Giver and Receiver. But I can't isolate those who are alts because I have no further info who among are alts 
• Exclude people who've never given any merit - It is already excluded

Thank you for your suggestions.


Thank you for your appreciation.


Thank you for appreciating my posts.


Nice suggestion, if the page view per user data is available that could be done.

Those merit points sent to Theymos during the early days after the launch of merit system, when lots of users didn't know how to deal with their sMerits, and part of them sent sMerits to Theymos as a way to show their respect to the forum admin from launching such amazing, helpful tool for the forum users.
By the way, you are likely that a big fan of "The Prison Break" film. ^^

most of merit has been gone to they mos lol good job bitcointalk comunity

That's is certainly a factor. You can look at the Google search stats to see there is always more interest during a bull run, the same thing happened in 2013. I don't think it possible to put a definitive split as to how much effect that had on sign-ups versus how much effect putting account farmers out of business had. There were people farming hundreds of accounts at the same time. It took a year and a half to build an account up to Hero when they could sell it for several hundred dollars. These people were signing up tens of accounts every week each.

I think it has nothing to do with Merit. It is due to the record high of $19K  USD price of bitcoin.
If you all recall correctly "Bitterex"  suspended account creations and people are  selling bitterex account at 1 ETH to 1 BTC . Binance suspended for a couple of week.
Cryptopia suspended Doge and LTC trading and other exchanges are also going offline as they are not able to handle the traffic.

The introduction of the merit system has to some extent helped maintain decorum in the forum. Looking at your analysis, the reduction of the influx of new members only tells one thing. The inability to rank up again based on post and activities counts weakened those creating new accounts because they know, that they will always remain in the same rank forever.

A lot of work has been done, you're done. From the article it is clear why some high ranks use their sMerits! In the end, sMerits will be spent and then what?

It's unfortunate that it falls onto the responsibility of the mods to create systems around campaign managers not performing their duties properly. It only harms the campaign when one person claims to be doing the work of multiple people. I'd think ICOs would have to address this issue over time if BTT didn't intervene, possibly by prompting a new branch of community services to develop ranging from self proclaimed bounty police to independent bounty auditing services.


This makes more sense to me now, thanks for the clarification!

I still think that while this may have had an impact on a few hundred, the biggest factor is bitcoin price in terms of new account creation. The late 2017 bullrun had a huge impact on the variables and we can't compare the results to today or any period after merit was introduced unless we experience a similar situation in the future, so I remain skeptical of the numbers that average half a bullrun pre-merit and a bearish market post-merit.

As an example, here's what I'm seeing;
Prior to merit - From 1/16/18 - 1/22/18 on the 'New Members per Day Chart' there are roughly 5000 new accounts created each day(super rough numbers) and the price of bitcoin was averaging 11.4K USD.
Post merit introduction - we're still seeing between 4000-4500 new account creations per day around price spikes of ~11K

I believe the actual impact is a lot less of a reduction than the suggested 36% average because of these variables and I'd like to see data with similar conditions. Could we possibly get these numbers with December/January eliminated to see the lower average result?

People should  understand  the benefit  of merit   :
For bounty hunters  less competition  so more rewards .
For the forum   reduce the spam and increase the  number of high quality  topics  .

So it's a win win game  for real people using the forum  while it's  bad for those that use dirty ways to achieve their goal ( $$ ) .

I am happy to see that newbies are also getting good number of merits. They need it the most. If I have to choose one post with equal weightage to merit, I will give it to newbie over an old member.

Regarding the merits for staffs, that figure appears very low. Does it include the merits they got on official thread like one related to merit system itself  ?

I wonder if my experience may be trypical. I started off with 80 sMerit to award, and I was fairly active in hunting for good posts, and I started quite a few threads about merits and posting. As a result of this, I was awarded 333 merits. and this gave me an additional 166 of sMerit to award. I spent a substantial amount of time in my quest to help the forum by awarding these merits. I must have got something right, because Vod gave me a positive trust rating for my efforts. However, recently I've felt myself becoming increasingly abrasive as a result of the difficulties presented by the current posting habits. I'm awarding very few merits now, although I still have some left in the bank. It's also some time since I received any merits, and I look on that as an indication of the fact that my posting now has a lower value for the forum. As a hero that doesn't promote bounties, I have everything I need in privileges, so merits are now not an incentive for me, but they are a recognition of public opinion about my posting, so I need to reconsider my attitudes here.

I have also read that several merit sources are not receiving their expected quotas of source merits, so perhaps Theymos is hoping that merit starvation will help to improve the quality of posting. It does seem to have had the reverse effect though, with many members frustrated by the difficulty in gaining merits, and making posts and starting threads in the hope of gaining a few. I suspect that we may see some more changes to the forum policies on posting, and I hope they will include the creation of some starter boards, and further control over the activities of the bounty spammers.

And many campaigns don't. That's the cause of a lot of the problems. All they care about is page impressions for their signatures and bumps for their threads. It is easier to run a bad campaign than a good one.

Higher ranked accounts can make more from their signatures so there were a lot of account farmers building up accounts to sell on mass. That at least is a lot more difficult now and probably does account for a fair bit of the reduction in new account creation.

The  Real evaluation of the merit feature would be after six months, However it is already obvious that merit is helping. Good thing exist because of bad things. many spammers get tagged and banned day by day after merit was implemented so it was really good. and base on my own evaluation, boards with shit threads are getting lesser. In reality, Don't give merit when the supposed to be receiver is not deserving, that's all. be fair and square.

Frankly, the work done with all these graphics is just amazing.
For me, it just lacks a graph with the places of the forum where merits are most given.
Let me explain, there is some local (where people speak little or bad English, like me), where the merits have all been given for over a month already and where it becomes difficult to win again .

Right, but many campaign managers still come across multiple accounts and ban them from participation as a result regardless of rank.

Why claim a bounty once when they can claim it 10 times or 100 times? Abuse of signature campaigns and bounties by multi-accounting has been going on for years here. Merit simply means that these new accounts will not be able to rank up.

Well, I have to disagree. We often forget about the local boards, in where there's a lack of Merit Sources, I'm afraid. Also, the decay in the merit distribution can discourage good posters to keep going with their amazing activity.
I think we need more Merit Sources, trustful ones (maybe the own mods) to encourage good posters to keep going. This is kind of unfair to make a real effort and can't rank up due to the lack of merits, either in your local board or in the English one.
The graphics are clear about that: the merit given-away are less and less every day, and even if it is based on the decreasing activity of farmers, the collateral damage is kind of dangerous, for it can discourage amazing posters to keep going.

I'm not sure if I quite believe that the merit system is reducing new account creation. While it would appear that way given the first chart provided, it seems to be a case of correlation not necessarily meaning causation. I believe the price of bitcoin itself is the only contributing factor to new accounts, especially because most newbies don't know about or understand the merit system. When you compare the peak dates on the merit implementation chart, they align rather well to bitcoin price spikes. We'd need another huge bull run with the merit system in place in order to come to the conclusion that it's clearly making a difference with new accounts. I don't necessarily mean to say it doesn't have an impact on the way people rank and the quality of posts from people who are looking to increase their ranks, but I don't believe it was implemented to impede account creation nor do I think it has a significant impact on that factor.


What would the goal have been for multiple accounts prior to the implementation of the merit system? If people were encouraged to post on a single account to rank up via activity alone, what would they have obtained from spreading the activity across multiple accounts? I wasn't around then so I'm not sure how the system may have been abused though there seems to be more potential in merit sharing across multiple accounts currently. I am missing some information to draw that conclusion so your input is appreciated. 

Definitly, yes. I think we are not too far away from that point though. Its very hard to earn a single merit now, just due to the fact that you need 2 merits to be able to spend 1, so available merits are slowly fading out (except merit sources, ofc.).

Thank you zentdex for another brilliant topic. Merit circulation continues to decrease, this is worrisome for people wanting to rank up. For me, this is the indication of bar raised too high. See how few are struggling to get Sr. Member rank. Most of the crowd are stuck on the road to Full Member. If the goal is to prevent keep as few as possible to reach high ranks, merit system doing its job pretty well. It cut the amount of account farming though, which is the good thing.


This quote is alarming. It's like when data scientist (zentdex) is telling to CEO(theymos) that something goes terribly wrong.

It is likely that merit system has made significant impacts on account farmers. Something like they stop attention to make new alt-accounts.
Furthermore, several account farms have been detected and got punishments from moderators. This is another good impact of merit sytem in terms of cleaning the forum a bit.

It is just a theory, but you maybe right.

For me, the graph before and after is excellent and you can notice that merit system is impeccable to prevent the spammer and alt accounts. Decreasing on the newly created users by 36% after merit's implementation within 100 days is very significant. We can say that merit system is really reducing the multiple accounts of people or spammers and focusing many members on creating a good quality post. On the last figure (ranking hindrance), you can see that most of the members need to obtain the required merit in order to rank up compared to those members that lack of activity if you use the old version of the ranking system. By the way overall excellent work zentdex.

Firstly, thanks zentdex for publishing such informative, helpful thread on merit system and its circulation in the forum almost 4 months after its start.
Secondly, I highly appreciate if zentdex publish those datasets which he used for those analyses.

Boxplots should be taken into consideration next time. 

I think that the decrease in merits receives is due a lot of the non-merit source users have exhausted their sMerits to give while some are still keeping their sMerits as like what you have said earning Merits is now pretty hard. During the first few weeks of the ongoing system we saw a lot of users sending 30 to 50 sMerits (which is the maximum a member can give to another user) to worthless post but the bright side of that is they become a suspect for cheating the system and was tagged immediately by some DT members.

Even with the lack of merits to give I do still think that there is no need to add more merit-sources for now as we still need more time for the merit system to do its thing which is to reduce spam and create more sensible post.

The tread is cool and all those graphics are just awesome! I guess you spent a lot of time for that. But the main point stays uncovered - merit trading and how to stay noticeable among all those other users with all kinds of posts? We can see on the example of this post that in the most of case just few first comments in the thread get merit, I guess it is because of users do not bother to read tons of remaining comments to the thread,  it is actually true,  well,  one more idea for a new graphic  

zentdex always creates an informative analysis regarding the merit system.
Some of his works that I like are the following;
10 Basic Info about Merit
3 Merit Stat After a Month of Merit
Merit Stat from theymos data
Where the Merit Pours?
I am pretty sure that this will be not the last. Thank you very much for all the information.
This is easy for us to understand the Merits system thoroughly. Keep the excellent work OP.

Someone who keep sticking on their constructive approach will have higher probability to earn merits in the future when spammers give up. As a result of this, constructive threads will have more chances to show up in eyes of higher ranked members, then get more chances to receive merits.
By now, constructive threads have been covered by shitty ones.

I believe that merit will not be (at least very minimum) distributed at some point/time after i see the chart of merit declining. It is like a natural selection from there. The forum will see which member who still contributing even if they dont get merit as much as he/she deserved, and who stays on doing it.

Average statistics are not enough, not exact to describe non-normal distributed variables.
Hence, if zentdex can share dataset which he used for the OP analysis, someone in the forum can use and help to make supplemental analysis.

Cloudflare was implemented about two months prior to the merit system being implemented. Cloudflare should have stopped the last of the bot accounts and made it more difficult to create spam accounts manually. A lot of the spam accounts won’t care about merit because their goal was to create back links to their website, not rank up.

Based on the DATA mentioned above, it really seems that the quality posters are also on the verge of decreasing, and with this numbers the Merits being distributed are also not that high... We might say that the system is effectively working with the standards of posting but the thing is some are now quite fond of not having Merits anymore.
It's absolutely amusing that the DATA stated above, that on the 1st month of this system was also an evidence that the Merits are being abused on that time, and now the difference is quite big more than ever. We already witnessed that this abusers are now having their punishments, getting tagged and  being permanently banned from this forum.
I completely appreciate this STATS from the OP. I hope that some important statistics from this forum will show up lately.

YES! Clearly the introduction of merit has reduced the number of alt or spam accounts being created. That's a huge win!

To zentedex, I have a request for additional analysis.

Is it possible to look at the average amount of merit given, per transaction? I'd like to see if there's a correlation between rank and the people who are gifting merit to friends or alt accounts.

Parameters likely need to include:

• Average of merit given per transaction, by rank level
• Being able to look at that average over shorter time frames would be useful (a week or 14 days maybe), the longer the duration the more likely the 1 merit transactions will dull the results, we're looking for spikes
• Can we isolate transactions by forum board? Some boards are more likely to have these questionable transactions (e.g., alternative currencies, local, etc.)
• We should isolate the received side of the transaction, an alt account is more likely to receive transactions that are arbitrary gifts than due to genuine quality of their posts - obviously just an unsupported hunch on my part
• Exclude people who've never given any merit

Great analysis, thank you!

I would be interested in the forums view on the decrease of merit over time and if that the current merit per day is around that envisioned at implementation. I understand the drop off immediately after the introductions as everyone was given a bunch and were throwing them around, with a large number of farmers merits going to their cattle accounts, but is it not at quite a low level. The fact that it is still in decline even though it has been around for a while is what interests me. In my opinion it shows either a lack of interest in handing out merits or a constant raising of standards, or possibly even a hoarding mentality due to the scarcity (which really has no benefit).
Are there enough merit sources? do they have enough love to spread around? Are they spreading it around? At what point in the plumbing is the trickle down effect being blocked?

I actually didn't see the give and take of merit that decreases each day that passes by as a problem. It is because this only shows how the current issues affects the system(well actually there are a lot of factors that affects the demand and supply of the merit)

Good numbers tho, because even if this is not the official data I think that the sources is reliable which means that the probability is actually high and reasonable.

It makes sense!
Campaigns which managed by famous managers like yahoo62278, Woshib, etc. seems to have better participants than others.

For the OP's creator,
As as mentioned in my previous threads (somewhere in the forum), due to unavailable detailed data given by the forum. Those sort of data analyses have significant drawbacks, at least in terms of its reliability.
In my opinion, only IP technicians/ experts can get such datasets from the forum. Theymos only given raw datasets.
Of course, I have no reason to defend the OP's creator, who made extremely good, helpful, nice-graphics analysis.
In addition, I think it might be better to give other users public datasets, on which the OP's creator used for the analysis.
Consequently, other members can verify the reliability of the analysis, and do further forensic analysis (if necessary).
I don't see any reason to keep those datasets in secret, because the main purpose of the topic is to contribute to the forum, right?
So, why not publicly publish datasets? Not only zentdex, but also others who published their analysis on merit system.

I would be interested to see more information about those who can’t rank up because of merit. I would be interested to see how long their posts are compared to others (specifically those who are one rank higher), where these accounts tend to post (what sections), if these people are participating in any kind of advertising campaigns, how much trust these people have sent/received.

The above information may give somewhat of an indication of how effective the merit system is working.

Op nice work even if I think we must consider a Δ on the value, based on my calculation the untrusted merit(abuse and exchanged by 50 from legendary to legendary for no reason, and believe me they are a lot) was around 40% last month.
The part that saddens me is that now we are in a situation where the abusers have ranked and for the big part they are unjustly unpunished (I've reported around 100 Merit abusers with, in some case over 200 sMerit and the eth address link, abuse and they are still unpunished).
Also now the monthly merit is very similar to the merit generated by the sources which shows that some of them are afk/don't care/(don't find post to reward, this is in my opinion an excuse demonstrated by the fact that some good sources like TMAN finds a lot of posts where to use the sMerit).
In my opinion now the montly merit without abuser is around 15000 while sources generates 17750.

What activity number are you using for heros to rank up?
775 or 1030 or somewhere in between?

You can't be using 1030 because you couldn't have gotten from 775 to 1030 since the merit system was implimented..

IIRC if you had 775 You got the 1k merits to be a legendary even if you didn't hit the legend lotto yet right?
I think I missed that by as close as possible..


I find it hard to believe that I am one of only 2 heros that just need merit to rank up.. That is quite an accomplishment..

I think their may be something in the legendary lottery thing that is messing with your charts? Or messing with my understanding of them..

100 days the merit system seems to have gone well with the graphic view, the number of members deductions is evidence of the decrease in spam accounts that just stay and go without the desire to become an active member and not just looking for $$$ but also trying to understand and recognize bitcoin, crypto and blockhcain

@ the pharmacist   they can  actually  use both things  .. enjoy bounties and grow their knowledge  about crypto  so that they can benefit their selfs from the $$ they got from bounties  as well as benefit other members  from their experience  and not just dump it in the first exchange  then  join another bounty and so on  ..  i think it's a shame to be in such a forum and never learn even small thing from it .

For the main  topic  it was intresting to know that  the more people give merit  the more they get back .

5152 new users daily before the merit system was introduced?  Holy shit!  Seems obvious to me that this doesn't represent distinct entities but rather just new accounts--multiple accounts of people.  I wonder how many separate people this actually reflects.  My guess is maybe 200 or so, if that.  Still, it's good to see that number dropped considerably post-merit introduction and hopefully it'll keep dropping.  I'm not saying new users is inherently a bad thing, but I think a lot of noobs are just here to participate in bounties, have zero interest in crypto or bitcointalk, and are going to be lazy shitposters.

Nice data presentation, OP.

Way too many graphs but take it as a good thing. People may see it in their on way and may be understand from a graph that really suits them the most. I do like the first one which is the before and after.
Good stats really.
I think the decrease of 36% from before is not a problem and I think that it is instead an improvement. The fact that your stats say that given merits are less means that merit abusers were lessen. We have noticed before that some users send 50 merits to a nonsense post and I think some of them are now very careful in sending their sMerits which is a good thing. They are somehow forced to find a bit of quality in a post.
Your efforts are appreciated. Keep up the good work.

100 Days of Merit

Hello, there fellow members! It’s been a long time since I look into merit stats.
In this post, I will cover the 100 days of merit. Just like before the data are based on theymos data which can be view
here https://bitcointalk.org/merit.txt.xz and other forum stats which can be view here https://bitcointalk.org/index.php?action=stats .


I. Before and After
Some are still wondering, do the merit system prevent people from creating dummy accounts?
100 days before the merit started the number of new users increases from a minimum of 1,662 to a maximum of 14,274 new users a day.
Fourteen thousand new user registered in a day is almost the same amount of new user for a span of 40 days a year later.
The figure below shows the 100 days before and after merit implementation.



The average new user a day before the merit is 5,152 this go down to 3,315 a day or 36% decrease after merit implementation.
Even though it decreases, we are still unsure if those new accounts are dummy or not.

- - -


II. Giver and Receiver
Merit is created to ensure a quality post contribution to the forum. Some member finds an excellent post and gives merit, while some give merit as a reward.
Some become merit sources and distribute to others, while some prefer to receive merit and hoard it.
There are 128,079 merits given from January 27 to May 7, with a total of 59,297 transactions. The figure below shows the unique number of accounts that give and receive merits.
Ranks and position are based on May 7 rankings. Some have ranked up within the 100 days but based on my estimate there could be less than 4,000 accounts that have ranked up.



The total shows that there are more people who receive merit than those who give. It could notice that most of the beneficiaries of merit are form the lower rank.


Color denotes the receiver of merit. The figure above shows Hero and Legendary don’t hesitate to give a vast amount of merit to Jr members.
Mostly, the higher the Rank, the more chances of receiving higher merits.


The figure above shows the frequency of giving merit to another. Most of the Legendary members give merit to another legendary, while most Hero gives merit to legendary as well.
Members give more frequent merit to another member, while full members, Sr. members, Jr. members, and even staff prioritize to give merit to Members.

- - -


III. Merit Decreases
After the merit was announced, most of the forum members give merit to overcome curiosity. But as time goes by, they realize that accumulating merit is not an easy task.
 One will need to receive two merits in order to give one merit to others. The figure below shows total merit sent per day from individual members.



From more than 4,500 merits given in a day, it dips below 1,000 after 41 days. And from 1,000 merits on March 9, it almost reaches 500 after another 40 days.
The fewest merit sent in a day is 523 happened last April 22 which is sent by 190 members.  Merit giver also decreases, from a maximum of 1,000 down to 190 per day.


Looking at the Merit Sent Moving Average, it almost decreases 1% every day. It would take an estimate of 252 Days before the average merit sent per day go below 100.


The figure above shows the amount of merit given in a single transaction. It could notice that as time goes by, the bulk of giving high merit also decreases.

- - -


IV. Additional Merit Stats
Merit Amount description


There is 41,265 merit transaction that gives one merit, 8,610 that give 2 merits, while 9,421 gives 3 or more merit.
If you question would be, why divide merit into three. That is because the average merit is 2.16.
That is why it could be a good indicator that once you have received one merit, it could be considered as low, 2 for average and 3 or more for high.   

Merit Average
Based on the table below it clearly shows that Newbie and Jr. Member have an average of 3 merits received in the span of 100 days.
It could be assumed that they could easily rank up to a member rank.


Merit Day & Time
Are you looking when the perfect day to post so that your will got a higher chance of getting merit is?
The figure below shows that most of the members give merit during Monday.



This figure shows the time most of the merit is given.


- - -


V. Hindrance of Ranking
There are a lot of members who complain that merit hinders them to reach a higher rank.
Thus, I tried to figure out who among can’t rank up due to merit, due to activity or both.
Table below shows ranks and their hindrance in ranking up.




The figure above shows that 45% of the members need to attain the required merit to rank up.
But it is only 4% higher to a much more more significant problem which is the combination of Activity and Merit.
While some, see merit as a problem, there are members who gain enough merit and just needed to gain additional activity to rank up.

_______________________________________________________________________________ __
Thank you for reading this long post.
I hope you have learned something about it. Once again, this is not an official forum stat.
 Some information may not jive as you want, but I am always welcome to comments and suggestions.