1. Introduction.Lately, I’ve seen multiple posts that hint or openly state that sMerit is only been sent to our buddies/friends/¿alts?/same rank/”want merit back from” type of accounts. Is this true?
When you open/read your (online) press during the day, just how many news sources do you actually consult? There are thousands of sources available, but we tend to read those that we enjoy most or are aligned to our views on matters. Once we work out what we like, we tend to make it a habit to read information from a fixed set of sources, and every now and then we may open-up to others and enlarge our circle.
We could take this down a level and state that, within our favoured sources, only a certain amount of sections catch our interest, and within these, articles and columnists.
I think that the above analogy applies to the Forum, were we move around certain sections that become of interest to us, there we favour reading certain topics, and we certainly get hooked to reading posts by specific users and favour their lecture, since we enjoy them, agree/disagree/agree to disagree with them, and ultimately find some interest to what they generally write.
The above conforms our general habit factor, and also relates to our network size factor. It doesn’t mean that we don’t open up to more sources within the forum, but as time goes by the speed of assimilating new members to our network is bound to decrease.
In many cases, our network expansion graph will probably look like a logarithmic function curve: Our network size grows quicker at the beginning, but tends to slow down (not stop) over time. Merit sources may be an exception.
2. First Approach: Merit sent by rank to rank.The first approach to resolve our initial question, about merit sent to buddies, would be to see how merit is being distributed from one rank to the other ranks.
We saw this exercise done in the post
Which Ranks send sMerit to which Ranks - and who ranked up .
It’s true that the post is now two months old, but it proved that ranks were not solely concentrating on meriting their own rank or above, but were in fact very active meriting posts belonging to users with a lower rank.
So users are not concentrated on meriting buddies on their own rank (or above), but are really of a broader nature.
3. Second Approach: Reciprocal MeritThis is the core of this exercise. If the opening statement were to be true, we should find that the vast majority of the users we send merit to also send us some back at some point (reciprocal), although in different quantities. If it is indeed true that we do send sMerit just to our buddies …
3.1 Global Reciprocal Information
The data is to be interpreted as follows:
Rank: Rank
nUsers: Number of users in the Rank that have sent sMerit to another user.
nUsersSent: Total number of users that the nUsers have sent sMerit to.
nUsersSentReciprocal: Number of users out of nUsersSent that have also awarded the nUsers with sMerit over time.
nUsersSentNonReciprocal: Number of users out of nUsersSent that have not been awarded back sMerit over time to the nUsers.
nMeritSent: Total amount of sMerit that the nUsers have sent to the nUsersSent.
nMeritSentReciprocal: Total sMerit sent to users that have awarded the nUsers with sMerit over time.
nMeritSentNonReciprocal: Total sMerit sent to users that have not awarded the nUsers with sMerit over time.
Network Size Reciprocal: % of our network that is merited and from whom we receive merit back at some point.
Network Size Non-Reciprocal: % of our network that is merited and from whom we don't receive merit back at some point.
Merit Sent Reciprocal: % of our sent sMerit that goes to reciprocal users.
Merit Sent Non-Reciprocal: % of our sent sMerit that goes to non-reciprocal users.
Notes:
1. By “meriting back” I am not implying that the merit is obtained by a prior merit event. The order of events is not analysed here, just the fact that, at some point in time, they have reciprocally merited each other.
2. What I’m counting here are transactions of the nature “User A sent to User B “ x sMerits, where User A is one of the nUsers and User B is one of the nUsersSent.
3. We could also do this exercise from the “Receivers” point of view, but I think it will complicate this even further so I have refrained from posting it.
What the above table shows us is that:a) On Average (but who is an average John Doe right?), we send sMerit to 17,14% of users of our network, who have at some point also awarded us with sMerit (reciprocal). Therefore, on average, 82,86% of the people we send sMerit to have not sent us any at any time.
b) In terms of Total sMerit, we sent 24,13% on average to out reciprocal network, and therefore 75,87% of our awarded sMerit goes to people in our network who have not awarded us with sMerit.
The above varies from rank to rank, but it goes to show that we tend to award way more to people that have not merited us at any time (non-reciprocal) yet that those who have (reciprocal).
3.2 Global Reciprocal Information – By Reciprocal Segment
Being on the global scale, each rank is averaged, therefore withing this average there will be users that have all their sMerit in a network that is 100% reciprocal, all the way down to those that have a 0% reciprocal Network.
I’ve broken it down in the above graph by 20% clusters. This means, for example, that there are 84 Sr. Members that have sent sMerit to people that at some point have all merited them back (100% reciprocal segment), 4 Sr. Members are in the situation of awarding merit to reciprocal users in the range of 80%..100% and so on.
What we are not seeing simultaneously is how much sMerit we’re talking about in each case since I would need to break down each segment further an complicate matters further more.
We can see that the vast majority of us are in the low quintal (0%..20%), and a fair share on the 100% value. These cases are the “fishy zone”, but are on the whole they are not very significant in terms of number of cases vs overall cases.
3.3 Global Reciprocal Information – By Reciprocal Segment and awarded sMerit Segment
The above is a break-down swapping Rank for sMerit Sent group. The average for the groups don’t differ too much from the global averages, although the [40..49] sMerits Sent group does stand out a bit more.
We can also see the extremes differ too for obvious reasons: The [1] sMerit Sent group is the lowest reciprocal group of all, followed by the [500+] sMerits Sent group.
3.4 Top sMerit SendersI’ve done the exercise of grabbing the 50 sMerit Senders of all times, and these are their ratios:
user_id name rank NuserSentReciprocalOverSent NMeritSentReciprocalOverSent
72795 QuestionAuthority Legendary 18% 24%
234771 suchmoon Legendary 11% 22%
98986 TMAN Hero Member 21% 52%
140584 EFS Staff 10% 18%
30747 Vod Legendary 20% 32%
153634 dbshck Staff 8% 10%
55384 Foxpup Legendary 11% 24%
1192397 paxmao Full Member 5% 12%
382413 xandry Staff 2% 2%
507936 DarkStar_ Legendary 18% 28%
290195 achow101 Staff 23% 56%
18321 OgNasty Donator 8% 13%
51173 mprep Global Moderator 11% 29%
24140 qwk Donator 17% 35%
23092 malevolent Staff 4% 3%
252510 JayJuanGee Legendary 22% 35%
569455 BobLawblaw Legendary 58% 91%
347141 BitRentX Staff 20% 29%
176777 mindrust Legendary 10% 9%
487418 The Pharmacist Legendary 23% 46%
459836 LoyceV Legendary 28% 36%
452769 bones261 Legendary 11% 20%
479624 Last of the V8s Hero Member 46% 77%
553678 rickbig41 Global Moderator 11% 13%
698159 Jet Cash Hero Member 16% 21%
976210 nullius Copper Member 54% 84%
113670 Mitchell Staff 18% 16%
988740 frodocooper Staff 4% 9%
520313 Lutpin Copper Member 36% 43%
379487 LFC_Bitcoin Copper Member 14% 36%
33156 vapourminer Legendary 3% 5%
120694 xhomerx10 Legendary 25% 47%
115423 Micio Legendary 42% 56%
255065 ebliever Legendary 0% 0%
346731 minerjones Copper Member 29% 27%
60820 DannyHamilton Legendary 29% 41%
112208 Vlad2Vlad Legendary 15% 12%
181801 shorena Copper Member 32% 36%
308793 1Referee Legendary 22% 16%
35 theymos Administrator 33% 48%
18312 phantastisch Staff 24% 9%
88912 600watt Legendary 32% 46%
35501 cAPSLOCK Legendary 35% 50%
129726 explorer Legendary 29% 51%
465017 actmyname Copper Member 19% 34%
211419 redsn0w Legendary 7% 6%
54791 Dabs Staff 10% 26%
389331 RHavar Legendary 43% 45%
511899 RoomBot Legendary 21% 32%
434984 mhanbostanci Legendary 12% 20%
I’ve added a full list of uses with their ratios here:
Bitcointalk Reciprocal aggregates by user 20180531.
I came to the site in early January 2018, and had never had any prior knowledge about it nor it’s members. My ratios are therefore natural organic, with no influence to pre-merit kick-off relationships since I had none.
My ratios are:
user_id name rank NuserSentReciprocalOverSent NMeritSentReciprocalOverSent
1582324 DdmrDdmr Full Member 37% 58%
What this means is that roughly 37% of the people I send sMerit to at some point have also sent it to me. That leaves a 63% that have not.
In terms of sent Merit, 58% of what I send does go to those members that at some point merited me, but 42% doesn't. I'm also far away from the averages but I believe that my network size is decreasing in terms of groth speed, and thus I tend to merit more my current network.
This seems natural to me and falls into the factors I stated in the introduction.
In conclusion: in general people do not just sMerit their buddies as shown by the averages; far from it.
It is true that one tend's to move in certain parts of the forum and reads certain topics that tend to be what one finds interesting. Those topics may have a larger presence of people that one merits and vice-versa over time, since the topics are of a common interest.
I would consider normal 50% ratios without much of an issue.
