Topic: LoyceV's Merit data analysis (full data since Jan. 24, 2018; not just 120 days) - page 23. (Read 37569 times)

You're right here, but data on the number of active posters is harder to get.
On the other hand, I can also imagine (let's call it a hunch) Russians are more active at Merit trading than Germans.

I'm glad it's appreciated. I still have more plans with this thread, but lack of time kicked in.

That seems like a good use for Merits, I only use them to boost my ego

I just wanted to challenge that assumption. theymos has been very clear that it is going to be like a fiat currency with QE on steroids. There is a constant supply so that it will not be a scarce commodity and he's monitoring the situation and will keep adding more until he's happy there is enough. In other words, I'm assuming the trend will not continue.


I don't honestly think it will hold the people who are here for the right reasons back. Looking at LoyceV's first set of data https://pastebin.com/01evJ1jN posted 6th of March (41 days after Merit started) there were 397 accounts with 41 or more Merit. That's probably more than the number of genuine Bitcoin enthusiasts here. I want to keep an eye on that number to see how things progress.

You are right, I already found a very informative thread by zentdex where the decrease is very well illustrated:


'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.'


He also mentions that, if this trend is continued, the Merit Sources are too few (although they were recently upped). My point here is, that even the current rate of distribution is too low and even if there are more Sources it wouldn't increase (what I think would be necessary) drastically but only level out. Right now, Merit is a scarce commodity that gets rarer and rarer.
Even if you post valuable content, which is a highly subjective matter, you don't rank up in the same time you would have in the past (even without spamming), because of this decline of available Merit/sMerit and therefore the increase in value.





I see what you are getting at, but isn't the (active) member count important rather than the number of posts or the total number of threads? Fact is in the Russian board way more  Merit gets distributed and unless they are that more members (by the same factor) it's a lot more Merit per person. I highly doubt that the Russians produce content that is this superior that it would merit the Merit (sorry, made me laugh  ).

But I forgot to thank you in my previous post, I highly appreciate this kind of analytical work. Would've given you more Merit if we weren't on a tight budget and it needs to last for some encouragement in the Beginner and Help section  

Let me know what you need. I might be able to pull it. I have a process that can fetch the merited posts so should be able to get some useful info from that. Until theymos bans me for DOSing the forum.

Not sure what the "competition" is but I'm fine with sharing.

How does this compare to the total number of posts per board? The Russian board has (in total) 7 times more posts than the German board. If recent posts have that same ratio, Germans are about as likely to receive Merit as Russians.

I hope that I can do something to help the forum with my skills on statistical analysis, but I can not get data from the forum. Raw data is not enough to analyze, and I don't have skills to get more detailed data from the forum.
Someone has done something and got more detailed datasets than raw ones published by Theymos, but they don't really want to share.
It's a competition. I accept that fact. I will learn and someday I can get data from the forum by myself.

The average is heavily weighted towards what happened right after the system was introduced and most of the distributed sMerit that everyone received was used in the first few days. Since then theymos has increased the number of sources from 35 > 57 > 77 and also increased the merit allocation for each source at least twice. Many of those new sources are indeed active on local boards. All you are missing is that it is a problem that is being monitored and dealt with by the administrators already.

What about the local boards? Those should be the figures for local merit distribution. But how will this progress? Being active on the German board, I'm tempted to learn Russian  

Right now, there are 77 Merit sources generating ~588 merit/day
But in contrast to that, since implementation, ~2152 merit/day were sent. So unless the number of merit sources gets almost quadrupled, it get's harder to earn merit, especially in local boards. Or am I missing something?

It hasn't changed much over time. The percentage of merits going to thread-starters was about 31% in the first week and 26% recently:

I don't know (yet?); it may become more obvious when I have stats per board. Just apply, theymos is looking for local sources in particular.

That's probably the reason why so many Merit-threads have been created: it is (was?) an easy way to quickly earn a few Merit.
I haven't seen shitposters earn Merit with a low quality thread.

I think the red is still way too large. An average thread is probably far longer than 4 posts so more than 1/4 of all merits going to thread-starters may be giving the wrong incentive to shitposters.

I'll try to plot this over time.

Thank you for doing this! I expected the red to be larger than the blue, so this is a surprise.

All the more reason why I'm surprised
It would be interesting to see how this ratio differs per board, I can imagine the Technical section has much more Merit for good answers, and the Meta section for topics. It's all on my Todo list (and I'm still recovering from the flu).

@BTCforJoe: thanks for testing, OP updated.

No need. Your top post is well worthy of [at least] 2 Merits.

I just double-clicked the send button on the Merit page. I can confirm that double-clicking does indeed double-merit.



See you in next week's stats!

Since I'm digging through the data anyways... - https://pastebin.com/Cf8PDFAg

This is a list of first posts in merited threads. The ones marked with "off limits" are either deleted posts or posts in "privileged" areas (e.g. Ivory Tower) that require to be logged in. Can't vouch for 100% accuracy but I spot checked a few and it looked ok. So if the "msg" field in the merit export matches one of those then it's the first post in the thread.

Based on that:



To be fair, this is heavily skewed due to people meriting theymos' merit announcement thread, satoshi's thread, HODL thread etc.

Thanks, LoyceV. I don't have skills to do those things (getting data from raw data published by Theymos) because I am not IT technician. I simply have ability to analyze data; if you get more detailed datasets which contains more variables, I can help you. If you interested, I can help you.

More variables such as ranks of merit senders/receivers; total number of  rewarded merit points for each board; total number of merits (received/ sent) for each users. It's for new-comers in crypto ecosystem.

This is a bit tricky, as I can't have real-time Merit data. I'm still trying to figure out the best way to determine how many "free" Merits an account had received.


I'm also curious to see what percentage of Merit is given to the first post of threads, and what percentage to posts inside threads.

I'm currently too ill to do anything. What more data do you want?

I can help you to do statistical analysis for your objectives if I can get datasets. I am not familiar (actually don't know how to get data from the forum). Raw dataset contains very little information. We need more variables to do analysis.