Sample Data comparison
Let me first make something abundantly clear. This is my blog. There are many like it, but this one is mine.
My blog is my home. In here, there is only one rule that goes.
Mine.
You are welcome to my home. However, you must show respect. This is not optional. If you do not show respect, and behave like an a'hole you are not welcome here. I will block you, mute you, and will not interact with you. I do not care what you think or do. You can go to hell! I write and interact with hive blockchain on my personal terms,
I do not have a project, I don't want anything. I don't want your money and I don't want your time. I am here to do community service with my stake and my time. I am financially free.
Generated by AI, DALLE-3;
Prompt: this is my weapon, there are many like it, but this one is mine western movie digital art
Ok, with that disclaimer out of the way, let we look at some 'active' Leo user data, shall we begin? Yesterday I published this post below, which had quite a lot of interactions.
https://peakd.com/hive-133987/@azircon/results-from-a-single-leo-thread
Here, I chose a random sample of a single leo thread, and extracted a list of unique users who commented on in. I plotted their hive and leo assets on a histogram, and scatter plot.
Aparently Leo people, there aren't too many, only Khal and perhaps Nomad, who commented as leo.voter got very uncomfortable. This is spammed into my blog many times. After that I blocked and muted the account for spam.
- I DID NOT cherry picked 125 random account. I picked a random thread, which is what you should do in conventional statistical analysis.
- I am not tying to represent LEO holders. I am trying to test my hyphothesis, which is
~57% of the 'active' Leo users hold less than 1 hive or 20 Leo worth of assets
This is my blog, and can have any hypothesis, and test that with any data. It's my choice. I am not asking for your permission Khal. Then he says:
I already answered, only I will add mind your own business Khal!
Then he adds:
Why do I care how much hive power Leo active users have? I am not interested in that data. I am interested in knowing ....again....so that it gets into the thick head of yours..
how many 'active' leo users hold less than 1 hive worth of assets
It is the opposite of what you are showing. Also thank you for comparing Leothreads with hospital. I agree with you there. It is a place of sick people and AI spam. I don't want to go there!
If you create a dashboard with all the data, I will say that will likely prove my hypothesis:
~57% of the 'active' Leo users hold less than 1 hive or 20 Leo worth of assets
Anyways, lets ignore Khal from our discussion for now. He is concerned and running around like headless chicken because he can't get the DHF funding that he was expecting! No SHIT!
But let me ask, I am a regular lowly hiver. I write when I feel like, I don't have any project, and I am not a witness. I am as average as it gets, so why an (self-proclaimed) APEX LION like Khal, who is ..... wait for it.....
Investor and business operator. Founder of the @inleo ecosystem. Our mission is to build a thriving creator economy centered around digital ownership, tokenization and communities
I wonder why such a busy guy, sorry enterprenuer, with such credential, was spamming my lowly blog? And then had time to make a 2 hours video of twitter (the shitpost heaven!) calling me stupid!
I hope he doesn't have daddy problems, because it sure look and sound like one!
Alright, I digress!
Histogram plot of two threads
My previous post looked into this random thread (125 accounts)
https://peakd.com/hive-167922/@leothreads/leothread-2025-02-02-12-35
Now I will compare it with this one (116 accounts)
https://peakd.com/hive-167922/@leothreads/leothread-2025-01-31-11-39
When I combine them, I get 195 unique accounts. 116 accounts hold less than 1 hive :) Or 59%. So when data is combined the ratio is getting worse. However, this maybe a sampling bias, at least I hope so. It will taper towards 57% I hope! Point is, lets say the real number is 50%, is that good? No! It's dismal. That is my point.
I already explained how I got this data at my previous post, so no need to repeat it here again.
I am going to argue that I do not see any visual difference between these two plots. I will show you the x-y scatterplot in a second and there you will see the same pattern. What I am trying to argue is that the behavior of these 'active' users are so predictable that there will not be any variation of pattern between their data when plotted consistently.
One of the argument against it is, the sample size is too small. Really? I have discussed on this post where I plotted 4349 active hive accounts, and their HP went down to 500 HP. I consider that HP to be too low to make any difference in the major activities in the blockchain. I also argued that that is probably about ~50% of all active hive users. So if, meaningingful total hive users are about 4000, then 400 user represents 10% of the population a very large sample size for any statistical analysis. If I do two of the threads and get about 200 (195 to be exact) unique accounts that is 5% of the population I care for. To me that is a decent sample size to predict the behavior of the population.
You don't agree? Don't. I don't care. Go do your own analysis, I can assure you that it will support my hypothesis.
What is that?
~57% of the 'active' Leo users hold less than 1 hive or 20 Leo worth of asset
I got to repeat it over and over again as Leo people are juvenile, and kids learn by repetition.
| Feb 3 Post, LeoThread 2025-02-02 12:35 | Feb 5 Post, LeoThread 2025-01-31 11:39 |
|---|---|
 |  |
 |  |
I don't have to tell you that they look almost identical. If you think about what leo threads are, it should make sense. This pattern is what we expect.
- In Hive-Leo scatter plot, we should always expect 3 clusters.
- The guy I call neutron star, comments on every thread
- The regular folks, they are both in hive and leo, they like to interact
- The new users, threaders
- In Time - Leo plot, the threaders are showing up in a cluster around a recent time
- They will have no or little investment
This pattern will repeat over and over again, if you open any threads and all threads. In my mind this pattern is not the problem. The problem is, how little assets these threaders hold. They are about 57% and they hold no assets. Also this clustering pattern. I can potentially demonstrate that the time of the cluster will slowly change, but will always be at a narrow time for a particular thread.
So we are creating accounts. Making traffic. But not getting investments. Also we are not getting content creators. Do I care about how many account leo created? I DON'T.
Do I care about how much total hive active Leo community owners own? I DON'T.
As I said earlier, if a hive account holds less than 500 HP, its contribution to the blockchain is negligible. 500 HP = 10,000 LEO :) Yes, Leo is WORTHLESS.
Can I filter how many 'active' Leo accounts holds more than 10K Leo? I sure can. I will. That will be a different post.
I will see how many I get and I will find out about their hive etiquette:)
What good is a new account if that account doesn't invest and contribute to the ecosystem positively?
That is the question I will leave you with.
Last I checked, if you took every user on-boarded by leo.voter in the last 12 months, there is less than 1,300 HP powered up total. While this is pretty bad, it's no surprise as most users don't power up and just earn rewards or do nothing at all.
What is more interesting to me is there are 13,614 monthly active users added to Hive but only 1,797 users who have created more than 4 comment/posts that have been on-boarded by Leo. If you look at those who created over 10 comment/posts, there are 1112 users. If you drill down to anyone who has done over 1, the number is 3,457 and if you say anyone who has done at least 1 you are looking at 8,133. We still haven't hit 13,614, so I am not sure how you can represent an active user who hasn't posted or commented.
If you look at transfers from these same users, it's about 8600 unique, but to the user is around 13,600 which is where I think they got their metric from. As you know every user on-boarded get spam messages upon on-boarding from executiveboard and witnesses and dhf projects looking for votes.
That's really helpful, thank you.
Few days back, I did a query for all hive account to find 'active' users. We can debate about what 'active' users mean, but I wanted to be generous, and used this criteria.
I covered the results on this post. The query returned 4349 accounts which I plotted. I think this could be about 50% of hive accounts that 'matters' somewhat.
I would say 10 comments is a very generous bar (its curious that you also used 10). Using your analysis we get:
1112/13614 -> 8%; or 92% of the new accounts are essentially ZIP!
Why in our right mind we want to support such a project! This is in fact far worse that I thought. Reality often is that way!
I chose 10 based on the one year time frame as I felt that was the least (despite fairly insignificant) amount I’d consider someone is even remotely active.
Agree, any 'normal' human user who is even remotely interested in social interaction will likely be doing way more than 10 comments, perhaps even try a post or two. Especially there is a hope of earning here. We have a lot of experience observing this behavior for 8 years.
When you consider they have this going for more than a year now, I wonder if this is total sham! Nomad talks about time, and say no one comes in and becomes an investor on day one. True. But they typically interact. Given enough time, I'd say 1 year is lot of time, there should be thousands of accounts that have more than a month or two time, and if they are making less than 10 comments so far, I highly doubt they are human at all.
I don’t believe it is a total sham but the numbers are bullshit and the value isn’t even remotely there.
Agree, they are just clueless on how content based social media works, which is a shame as they are here long enough.
I forgot the new haf the table is moved out to balances out of accounts, so I was looking at rewards. Instead of 1300 HP, it should be this for all users onboarded in the last 12 months.
Top users being:
wait... 11 users in 12 months?
you like practical jokes a lot, so are you serious?
The top is the sum of all.
Below is top users that were onboarding in the last 12 months based on their HP (> 500 HP)
I don't joke around about data.
Okay, so HP is higher, but I never paid attention to that bit.
Is this bit still the same?
PS. but it is only 11 users that hold more than 500 HP. We can call them 'effective contributor' or whatever fancy word we like to use.
all the user counts are accurate, the problem was balances was removed from accounts table in the new haf but there are hive/hbd fields which are actually rewards not balances. It's kind of weird. I am in the middle of moving all my queries over.
I've stopped looking at Threads for now as it had low value for me. I don't care about 'AI' generated images. I use Ecency on my phone, but Waves is not great either. I don't care how many 'points' you got today. Snaps is generally better.
I know the Leo people care about what they do, but I wonder how effective it is. If they really recruited so many users then why do I see no real change in the number of Brits? I run queries for UK locations and find very few new ones. There are some created by leo.voter, but they are not doing anything. Inactive accounts are not helpful. I may add details on who created the accounts to my stats page.
I'm all for Hive growth, but boasts need to be backed up with data.
Takes cover...
No need to take cover. We are the stakeholders. They are begging for our vote. We decide what to do, not them. If they are crooks like I have proven they are they deserve your boot.
Having a bigger stake does bring some pressure to use it well and people may ask for votes on various things. I will make my own decisions on that.
They asked for my vote. I did not give it to them.
Obviously you will make your own decision
snaps is better, but there is also not too much interaction with other users.
There is some, but it may only be a few people really using it.
There is this initiative to share links on reddit to attract more people to hive where I take part.
https://peakd.com/hive-197333/@x-rain/hiveposh-contest-15
If you look at these stats there is a great difference between Hive/snaps and reddit.
https://peakd.com/hive-197333/@blkchn/re-blkchn-sr9jrw
On reddit users respond, discuss. Here on Hive is very low interaction.
Users are busy writing posting, hoping to get rewards.
On comments you do not get many rewards, so they concentrate on postings.
Thats my humble opinion.
I've heard someone else complain that Hive gets less interaction than other platforms, despite the rewards. We have far fewer users than reddit/FB/Twtr, so you will not find many people interested in a given topic. The simple fact is that we need far more users to get that interaction.
yes, it's the question of the critical mass.
If you reach it, you get more attraction, if not then there is a problem.
Can't get users due to lack of interaction.
Can't get interaction due to lack of users.
Can't win :)
Need pioneers to break the cycle.
When one sees a rat in their kitchen one does not feed it more food.
😃
Yeah if you can’t kill the rat or are allergic to rat poison!
I will still look for some Elon Musk branded organic rat poison though! :)
It will have his face on it and you won't know whether it is poison or one of his other products 🤣🤣
Soon, he will get a government contract of 'undisclosed' amount to manufacture solar power dildos! :)
Well they do say stick to what you know and that man knows dildos!
😁
Have we figured out what it means to cherry pick and pick random at the same time? That's some multiverse of madness.
I am beginning to suspect that they lack basic education! I am not saying they didn’t go to school, but saying they didn’t learn anything.
My assumption: randomly pick and lift to the 'cherry on top of a cake' status? It does not necessarily must mean that, but sorta hints, gives a promise that it may mean so. 😉
Fun fact, when I get shakes from Chik Fila my son will always steal the cherry off mine. Maybe that’s what he is referring to?
😄😄😄
sup!
@azircon What a great analysis, thanks for doing that! The reason why I love to read those posts is that first of all I can see everything what I think and see in my own eyes but with real and confirmed statistics. Secondly I'm able to find new tools like beebalanced , this give me an idea of how to interact with some users .
That is exactly the point. That is the reason I am spending massive amount of time to demonstrate what can be possible when you plot data that can be made freely available. Then you can verify any claim anyone is making. Crypto space is full of people trying to exploit you.
Totally right, thanks for doing that. We should actually make a descriptive guidelines base on KE Ratio - I see the guy who used to be pretty big blogger with KE Ration - 1048915 and on the page should be - "That person is not worthwhile; he probably took as much as he could from the system and then disappeared into the void."
Interesting data but most importantly, I enjoy the Full Metal Jacket references :)
Tbf, I've never looked at Threads, barely used Snaps to post Car Photos and I am not sure what I expected from those "offshoots" of the various platforms 🤔
Thank you. I enjoy full metal jacket a lot. There is a lot to learn from it.
Threads are short form content like twitter, and just like twitter it’s mostly spam. There is nothing to look there. I am just pulling accounts that use it as a sample
Some months ago I also had some data from Inleo, more specifically curation... despite of having some type of fun, since I like stats and data as well and saw a type of outreach, it was momentum. You and Marky have been exploring that in the last couple of weeks which brings more of this data, you probably had some fun like me testing a hypothesis and it showed the results that you were expecting and probably more interesting data that could generate maybe more digging.
My question is: Since I see mostly the same users commenting and arguing how bad this project became in the last years, right now their proposal isn't approved,it is below the return proposal, but my fear is that the sea will calm down soon, this post and the last one will get into the void of Hive (probably the real black hole isn't a user but our post archive), that's what usually happens. How can we keep this idea or motive alive? making regular posts about updates, perhaps? This question I ask myself every day so maybe other people can help me.
I hope some people start to understand this point. Making hundreds and thousands of accounts is completely meaningless if those accounts see no usage (or only AI usage). We need to get people here who will invest, who will write posts (real posts) and contribute.
Someone going to prove at some point that there is no human behind to thousands of accounts. Looking at the account creation patterns I suspect it is them who are creating them, but it is a hunch at this point. There are many smart people at hive, I am sure they will figure out this lunacy soon enough.
For once I feel rich. thank you.
This is funny- just yesterday I said the same thing.
Yes I saw your other comment. My sentiment is the same that I said there.
We have an ethical duty to treat hive properly because if we don't then we lose this pretty quickly
By the way do you think that it is better to buy more splinterlands cards and less SPS or rent cards and buy more SPS?
Right now probably nothing and hold cash.
I saw some posts taking dunks on Leo but I hadn't dug deep into it... The truth is that to a lame man it doesn't look too promising. For now, and I do say for now, I've removed my support for the proposal.
I'm willing to give the benefit of the doubt to anyone, more so when I see them working on a project for so long. But, these facts cannot be hidden easily, this is precisely why blockchain governance was created to begin with.
When I first got on here, almost 8 years ago, I was convinced hive (then steem) was going to kill traditional (can I use this term now) social media, but it's proven to not be the case, not by a mile.
That being said, I continue to be here because I do think that one day we will find our North, and recently VSC has me thinking of green candles.
Thanks for sharing this...
You are welcome.
More I dig deep into Leo onboarding data more dismal it looks!
Great work!
Would be interesting, how a random snap-container would compare! I bet it´s users would have a more healthy HP distribution.
But to some extent comment farmers and shitposters are attracted by those microblogging formats, aren´t they? A quick pic and some words, and ready to grab some cents of rewards. Still, on snaps it looks way better and more organic.
Pretty funny. I don't think they understand what "random" or "cherry pick" actually mean. While not mutually exclusive, the one often precludes the other. It's interesting they assume you had a position before you pulled the random data versus the fact that the data you pulled forced the assumption. Like you said typical scientific method, but since the data doesn't fit their narrative, you must be evil...
Most people do not have a coherent understanding of random samples, and random sampling.
Populations of objects can be classified into decision units, and then a sampling methodology can be devised to reduce bias from being introduced.
After I was a soldier, I worked as an environmental consultant and routinely collected samples of soil, groundwater, and hazardous materials (lead, asbestos, arsenic) - while the sample collection procedures are specific to the field or medium, the statistical principles apply broadly to assessing some portion of a population of any objects to measure any arbitrary characteristics.
A meaningful sample size has some proportional relationship, or ratio to the population size that we are attempting to assess. We may not like the inferences from the data but the data can certainly be collected in an unbiased reproducible manner. With hazardous materials some sampling protocols even mandated duplicate samples as a quality check for the laboratory.
Tangential, but this brings back memories of rappelling down a hillside in Hawaii to collect samples to measure the lead content in surface soils after a contractor made a mess from sandblasting a bridge coated with lead-based paint.
Good times.
Azircon's data may prompt some uncomfortable discussions or reflections. His value judgments on that data indicate what he considers important characteristics in Hive users - original content creation or patterns of investment or accumulation of core Hive assets. There is a lot to reflect on there, and it likely is an iteration of the ongoing debate about what makes Hive valuable and how is that value best sustained.
Yes. Khal’s entitlement and arrogance is bothersome.
I have blocked him.
I can't say I blame you.
Users? They are not users, they are farmers!
a large number of 'them' 'it' are just a fake name on the blockchain.
Interesting data and conversations:) well, that fits perfect with my analysis earlier the month. And I'm hundred percent by you that we need users that invest instead of leeching the system. I wonder if Inleo Ai Content Spam shit will ever rank on Google. That will be worth a post as well. Anyway thanks for the details. Greetings from the road back to Germany from beautiful Barcelona.
You are welcome.
https://peakd.com/@khaleelkazi/sporog
Here is another example of similar context.
Similarly when silverstakeruk requested for explaination about leo missing payments.
https://peakd.com/hive/@spinvest/spinvests-largest-hp-delegation-time-for-a-change
And other similar responses in marky comment section.
When community mentioned about node being down, it was also rephrased as an attack on them.
The scope of the proposal is to bring active users, not investors. Investing comes with time, nobody arrives on any ecosystem and buys in on day one. Changing the goal posts to draw attention to factors that are not the scope of the proposal to shine a negative light on something is intelligent yet sneaky, but simply incorrect.
Again, this comes with time. If we want investors and positive contributors, we first need to bring in users. No new users = No new investors. Simple.
Irrelevant as well.
Ad hominem arguments do not take away any validity to my point.
Get lost from my blog...Please.
Next time I won't say please.
So, accurately reporting your acts is an ad hominem? Good to know.
He is doing so to undermine my argument.
If the facts undermine your argument, it's your argument that's the fallacy.
🤣 ok bro
Maybe one day we will have AI generated active users here :)
!BBH
Being a Krill asking for advice... Shall I return to Peakd... I guess u will say it's up to me... I mean, they've got visual marketing campaigns and have recruited a ton of #Splinterlands content creators... I am a bit lost.
I use peakD for all my blogging. 100%.
I am looking forward to that dashboard with detailed statistics which was promised to be launched yestardy. Any news about it perhaps?
Anyway, thanks for the analysis :) I prefer data over drama :)
As to cherry picking or not... I went through the same procedure for a different thread. (Couldn't get leo balances though, its not in the online version of @beaker007's tool, right?). Took all commenters on it, and I get similar results for their hive/hp balances. This checks out.
Obviously some thread-to-thread variation. The one I tried had 40% users with less than 1 hive+hp.
Yes beaker is working on it. This is not paid work you know, and we have day jobs.
Yea, I was just curious if I was being stupid or not, I often am.
Hi @azircon,
Your post really got me thinking! 😮 I agree with you. it’s important for people on Hive to actually invest in the platform, not just create accounts and leave them empty. If someone’s not adding value, what’s the point, right? Though some factor might contribute to inactiveness, like some genuine post from newbies don't get reactions at all, that discouraged some.
I also think it’s cool how you used data to show what’s really going on. Numbers don’t lie! It’s kind of sad to see so many accounts with almost nothing in them. Like, why even join if you’re not going to contribute?
But here’s my take: maybe some people just don’t know how to get started. Maybe they’re scared or confused about how Hive works.
Anyway, thanks for sharing your thoughts. It’s nice to see someone care so much about Hive’s future. Keep up the great work! 👍
!BEER
Bots. AI. Not only worthless to social media, but a pox upon our houses. Leo is attempting to insert AI accounts to mine the human users left to profit the owners of the AI socks. If allowed they'll turn Hive into a parody of social media of bots talking to bots.
Thanks!
That is the awareness I am trying to raise. Agree 100%.
!PIZZA