Looking for a good Bidbot mathematician
I just haven't been able to get these numbers out of my head.
2018-03-12
total votes 748,848
paid votes 11,604
paid vote percentage 1.55%
2018-03-12 this was created by the blockchain for rewards
STEEM 47698
SBD 47958.43
source
2018-03-12 was sent to the known bidbots
7,328 STEEM
47,245 SBD
source
I am still trying to work out what this means. So, 700 more SBD and 40,000 more STEEM was created than sent to bidbots. Those bots than voted more than what was sent and collected curation. Okay, got that. You are going to have to bear with me here tbh as i have very little idea what I am doing with this.
So, they are going to vote some percentage more than they were paid right? 1.5x multiplier perhaps? Let's work with that.
So, ~10,500 STEEM value and ~71,000 in SBD value voted.
But of course that can't work that way. The vote is in SBD right? so if I convert the STEEM to SBD that will be ~$2 so, ~21,000 SBD. So the votes the bots sent out were 92,000 SBD worth on 2018-03-12 ??
Of course, that will get paid in a 50/50 Split so the SBD is calculated at a dollar. making it 46,000 SBD and 46,000 in STEEM if Steem is valued at 2, that is 23,000 Steem. Yes?
So on the 12th of March, before curation, the bots voted 46,000 SBD / 23,000. After curation (20%), that would be ~39,000/19,000.
But, the total pool paid was SBD 47958.43 / STEEM 47698
Now, I don't know how all the algorithms work and I would love if someone with actual math skills can do some math on this but, this is crazy even if quite wrong, right?
So, for the 98.5 percent of votes that weren't paid for, essentially out of the pool that day, they were only paid from the equivalent of 24,000 Steem?
From that, take away the App votes, curation trail votes, the circle jerks and self-votes, How many organic votes were cast and what value did they have access to?
I know that this isn't a very accurate calculation and not anywhere close to the way it should be calculated but, no wonder small accounts are struggling and people's votes aren't worth much. That is a massive amount of the pool being locked up by 1.55% on that day, isn't it??
Total votes 748,848-11,604 = 737244 unpaid votes had access to 24,000 STEEM
Paid votes 11,604 had access to the equivalent of 46,000 SBD and 23,000 STEEM
Is that somewhat close? I have to be missing something don't I? That means the bidbot owners/operators/delegators are not only getting the 7,328 STEEM and 47,245 SBD liquid, they are getting access to the curation of about 20% of 75% of the daily pool. And the rest of that 75% is being concentrated on 1.55% of the total vote pool. Yes??
What am I missing... please it is driving me insane...
That is ~50% of the pools vote value paid upfront and liquid plus 15% of the daily pool to the operators in curation and 60% of the total daily pool to 1.55% of the buyers. The remaining 25% is spread to the other 98.5% of votes.
Hang on... what? like... WTF?? Can someone please, please, please do this math for one day on the blockchain and put me out of my misery???
If this is the case, it means there is absolutely no hope for anyone else to grow organically considering the competition involved and the fact that NO ONE IS VOTING ORGANICALLY. At least, not enough with enough pull on the pool to make a significant difference.
Where are all the large private voters gone? What happened to finding content? How is this just 'the blockchain allows it' so it is okay?? What the hell am I missing here?? Even if my numbers are 50% off, it is still crazy.
Come on, there must be someone who has the skills to do the actual math here for me, for you...for us. Please? What if I send you some SBD, will you do the math then?? I don't mind admitting that my math is bad or that I have missed how the pool is distributed but, what?
@paulag @miniature-tiger @abh12345 ?? What will it take to get these numbers even approximately right? Please, my math is shit and I don't understand the algorithms for the pool. If you want, you can DM me the numbers in steem.chat and remain anonymous if you want. What about a dead drop? or email me from a throwaway account?
A bit of help because I am starting to draw these numbers on the wall with marker and I can't keep blaming it on the baby...
Taraz
[ a Steemit original ]
Still working on details, but wanting to point out a typo:
The SP allocated for that day is 19583, not 47698. This might make the picture more grim though.
thanks, missed that. I am not going to reformulate it all no though so I will vote this to the top. There is something messed up though with all of this isn't there? It isn't just me being crazy?
The numbers would be hard to track with the few numbers you provided. You are making assumptions on a ratio over the bid amount to the amount of bid votes. There are many times that there is More Bid then the Total Vote Value...yes people are that freaking stupid. This is why it takes work to buy bids via a bid bot unless you just accept being screwed over and over.
I'm thinking someone with access to raw data (@abh12345) might be able to pull the numbers that would be needed for this.
The numbers to figure out how much of the reward pool is going toward paid bids you would need a few data sets.
One data point that isn't a factor in the total rewards that you maybe added into your calculations is the Amount Bid Bot Owners Were Paid. This number only allows a guess as to how much voting power was used. The amount also has nothing to do with the current reward pool as it's paid for with previous rewards.
With those 4 data points I would be able to give you pretty good numbers for the amount that is being taken out by the bid bot owners/delegators.
That's not as straight forward as it may seem. There are lot's of bots that aren't listed in that one place. One had to check their turnaround first to see if that can be neglected.
Getting the total vote value is heavily distorted by "sell your vote" features of MB/smartsteem/booster/ and others. Separating payed votes from organic ones for vote sellers is close to impossible IMO. MB has an API that could provide this information, others don't.
Wasn't thinking about the smartsteem style sell your vote. That makes the numbers even worse.
All those importantsold votes are not listed on the one place. Minnowbooster and smartsteem private votes make up a LARGE % of sold votes too. Yes Taraz there is someone who can work out the numbers for you in chat.
I think it's reasonable to average this out or provide a lower bound. If you look at the bots, you actually see that 1.5 multiplier is fairly common. Even if you set the multiplier to 1.3 or something you would get something useful. As @tarazkp mentions, it's not likely to matter too much. My preliminary calculations along this route are showing similar results to what is claimed in the post. But I'm double-checking things... Also I rather expect this state of affairs since the distribution of SP is concentrated towards bid bots and whales. But I will be claiming that bid bots spread out votes, so that'll be interesting to see if I can get data for that.... I think it would be instructive to see the distribution of payments as well.
One issue I thought of is that the whales didn't always use 100% of their vote each day. So by delegating the voting power this voting weight is now used each day, every day. This alone changes the entire distribution of the voting pool.
Just an example I know someone right now that hasn't been on in over a month with over 10k SP. This SP sitting idle helps the rest of the reward pool vs if this SP was delegated to a bot.
Hmm. That's an interesting thought. Seems like this person is missing out :P
You are right. I'd say
Idle > bot delegation > self vote
In terms of freeing the reward pool for others. Of course this is probably not something we can count on.
How much made on bot delegation. Hmmm... Oh dear, it's quite similar to a self vote after all. (They get half the amount at worst case, best case they can pick up an extra 25% from curation... Even more if the bid rounds are over filled).
Ouch. I need to re evaluate my life.
Running out the door now and didn't get to reread my comment. Hopefully on first writing it makes sense...normally I take a few minutes and read it a couple times and make edits.
Yeah, assumptions are all I have when it comes to this. Essentially, I see numbers that I can't intuit some vision of and my brain shuts down. All my brain is telling me is there is a lot going to a few and not much left for the many...
HI @Taraz, very very interesting question, I would love to take a further look a the data and math, will get on to it today
Thank you :)
If I could make a suggestion I would say, don't use anything I have written in the post as a starting point :D
Even estimate figures on this would give us a lot of insight into the system and opportunities to improve distribution.
okay I have read back your previous posts in detail and the posts from @penguinpablo and as you suggested, used none of it to do my workings. Instead I have taken a slightly different approach.
https://utopian.io/utopian-io/@paulag/how-much-of-the-rewards-pool-is-paid-out-to-bitbots-v-s-organic-votes
You know what, since your post i've been wonder just how much 'free float' is actually available each day too.
If you take out the bid-bots, utopian, dtube, dlive, steemit and other non-voting whales, what's left of the pool?
It could be a decent utopian-io analysis and I think the guys above or @crokkon / @lextenebris could do it justice too.
I was chatting to someone earlier about things on a more general level, and comments like 'you get what you get each day' and 'the chance of whale vote has gone' were some of what was said.
Did too much get pushed to the Dapps too soon? Is it delegations to Bots that's done it? Or is it a case of jerkus maximus + TA guy?
Gimme a pie chart someone :D
This is what I mean, I have no idea what is actually available for the other 98 odd percent. Even the basic numbers are well out of my skill range. Just to see what happens on 1 average day would give a huge amount of insight.
Many whales who used to curate are making more with less work through delegation, the dapps are often narrow in their spread or 'random' and there are several very obvious vote trading circles going around.
I feel like every time I post about this I am digging my future here deeper into a grave though...
Perhaps I should just post photos.
ha!
It's a popular road to the dark-side, but I wont link any 'community' giants :)
Someone will have a look at this though I'm sure, i'll start throwing the post at people tomorrow :)
I had a go at this yesterday too.
Final conclusion is that over a week there's 53 cents a post available.
That leaves nothing for comments, and probably a whole lot more.
How do you 'feel' about that number?
I think we should consider ourselves lucky to be doing a fair bit better.
It doesn't look that good for the new sign-up though, and I think it's a case of investing or focusing on one of the d-apps or utopian.
Of course the price can go up, which can make the number multiple quite quickly, but perhaps a higher reward % needs to go to curation to try and lure accounts away from bid-bot delegations.
The problem I see with more going to curation is that the bots will adjust, offer less but gain more in curation. Does it change anything other than take a little more away from content producers? I was very supportive of more to curators but, the major curators have given up on organic discovery (and voting almost entirely manually) which means new users and users not already gaining a lot of support are still unlikely to be found. That is my intuition on it anyway.
The goal would be to persuade bot delegators to to return their SP and go for curation. I'm not sure if switching the %'s could solve that in anyway though.
And yes, it doesn't change much for new undiscovered creators at all. Which is why @curie, @ocd, + other curation teams remain so important, and should, in my opinion, also receive some of that steemit dSP.
If you want a good comparison go to the T/A's page and then visit Andrustovius page. He is very good and is growing with around 250 followers already. He upvotes comments and if you want any info does the charts for you same day. Some cases tonight within a few minutes. He is not making $100's off his posts but cents. Just hope he grows but he is frustrated and he is the real deal. The site needs more of that type of person.
I've been wondering myself just what 44 million of parked SP (the steemit account's current stake) plus another 12.5 million parked SP across 14 accounts ranging from 341,000 to 2.2 million SP might mean as far as the reward pool goes.
In between those accounts are others who are mostly delegating their SP (so don't have a whole lot more than that to upvote with—if they chose to). And then you have everyone's favorite TA's whale buddy. So, misterdelegation and freedom are the ones primarily delegating their SP to either apps or bots. hendrikdegrote is delegating most notably to sweetsssj but he also does a lot of upvoting in the curie curation trail. The steem account creates new accounts and delegates SP to them. And then thejohalfiles just recently delegated most of theirs to smartsteem and utopian-io.
All told between parked and delegated SP, that accounts for around half of the steemitboard designated whales and a few of the larger orcas.
@paulag had an excellent analysis about the unused Steem and its potential impact on the pool: https://busy.org/@paulag/the-impact-of-unused-steempower-on-the-rewards-pool-blockchain-business-intelligence
Thank you for pointing this out to me. It always seems that someone else sometime has had the same question (maybe even multiple times).
I've read through the analysis she made and basically got that the reward pool empties out if even one person sitting on vast SP holdings decides to employ it.
Then, lextenebris counters in the longest comment I've ever seen (even long for him, which is awesome!), that basically, you can't drain the rewards pool, it just dilutes the worth of each upvote down the more people vote. Or something like that, since each vote, especially large votes, move the reward pool in different ways. At any rate, no draining.
But then freebornangel comes in redirecting the conversation back to where paulag's post began, with whether or not the rich here keep getting richer and the poor at best stagnate and how hypocritical everyone who suggests otherwise is.
So, if I can summarize completely, it's better that the vast reserves of SP don't do anything, including delegate, because then the rest of our upvotes are worth more, which I think affects what @tarazkp, this post of yours is about, right? :)
I think you haven't seen many @lextenebris comments yet. :D
But he is correct, you can't drain the pool as such, it's constantly replenished. As you correctly summarize, the more SP is actively used, the lower the upvote rewards per vest are.
So... let those sleeping whales/distributors continue their hibernation as long as possible, please. :)
So, the overwhelming majority of the ecosystem is pay-to-play? What a shame. If true, this information could signal a sinking ship due to a lack of organic voting. I suspect the majority would rather give away their content for free rather than pay to make money.
The whole thing reeks of a money vacuum, where the stupid give their money away to the investing class. The inverse robin-hood. This centralization of value makes the value inherent less valuable.
Well, there is an orthogonal means of interaction:
You can write things, post them, and then look around for other people who write things which you enjoy and might be related, engage them in comments, and literally grow your audience by attracting the attention of people who might be interested in what you do.
I can't promise it's profitable, but that part can be fairly emotionally rewarding.
Unfortunately, the real problem has nothing to do with the voting system or the fact that people would like to make money. There are problems with the former and the latter is just to be expected. No, the real problem is that Steemit as a social network is poorly designed, because it's missing a significant and important part of what has made social networks popular over the last decade.
Discovery.
I've written about this before and if you're really curious you can go through some of my previous work, but the short version is that there is no way to silo content, no way to sift through content, and the basic assumption is top-down authoritarian when it comes to how the system values content. What you vote for or I vote for really doesn't matter to what the system presents to us except in vast aggregate, and unless you agree with everyone else on the platform – the things which are surfaced by the platform for you to read are not in accordance with your interests and expressed desires.
Now, it's possible that after the next hardfork when very rudimentary community management systems are going to be put in place that we will see a really solid engagement with creating actual mechanically supported communities, which should make it much easier to find at least some content.
With the absence of a decent discovery mechanism, a reasonable community architecture (even the rough equivalent of sub credits would have been a vast step forward over what we have), or any means for your personal decisions to really impact what the system presents to you for consideration – there is very little left, from an architectural point of view.
If you just treat the platform as a reasonably useful blogging platform which conveniently uses Markdown and provides rudimentary account following mechanisms, and you look to off-site mechanical architectures to define and build your community (like Discord or even Facebook Pages), you can actually make some fairly significant use of Steemit as a social network. But it is essentially incomplete in and of itself.
You're definitely right about discovery being an important part of social networks that Steem does very poorly at. Even those "authoritarian" models give me options that I sometimes do enjoy consuming. I rarely get that feeling dumpster-diving on Steem. Though you can find interesting gems if you search long enough.
I don't really care about making money writing as much as I did in the past. I'm more interested in finding interesting content that satisfies whatever craving I have at the time. Also it's been interesting to see how much this platform has transformed since I first explored it in 2016.
Hopefully communities do reach us sometime in the near future. Like you said, by developing sub-communities it should easier for like-minded individuals to interact more with each other and provide each other with engaging and interesting content. Although, looking at the GitHub pages, it looks like Steemit is really only dedicating one developer to its development and recently reorganized and adopted the agile methodology. Hence the massive delay.
Even if the ship does indeed sink, the platform will still be interesting to use. Witnessing the rise and fall of a social media platform. That'll be something interesting to write about.
Let me be clear about what I mean when I say "authoritarian" model.
On the steem blockchain, the only measure of quality or readability is the accumulated weight of vests which are assigned to an individual piece of work by votes. Like everything else on the platform, it's scaled by the SP of the actor – in this case the voter.
You have complete autonomy when it comes to how you wish to pass around your voting power, which is really just a separate pool scaled by SP and which regenerates at about 2% an hour. But it really doesn't matter what you vote for, because the only thing that matters to the system as a whole is what those who have sufficient SP to make their votes far more valuable have voted on.
This wouldn't be a problem if we were talking about less than two orders of magnitude difference between the average user and the top end, but we're talking about seven or eight orders of magnitude (and often higher) between the average user and the top end. Ultimately, this means that what you and I decide to vote on and reward is not what the platform puts forward as the best content, for us individually or globally. Only a global rating makes a difference to surfacing, and we can see the results of that relatively trivially.
Look at Trending and Hot, and how utterly useless they are for actually discovering good content. Those are the only surfacing mechanisms that Steemit provides.
When I go looking for content on Steemit, it is hard work. There are a limited circle of people whose work I consistently look at, but mainly it comes down to picking a handful of tags which aren't so omnipresent as to be useless and not so rare as to turn up nothing.
Depending on the relatively small number of people that I follow to re-steem content that I'm likely to want to see is also in my quiver, but the content worth looking at is rare enough that even that wider net doesn't always find it.
Thus, on top of those mechanisms, I have to go off-site for tools which actually serve my purpose. Ginabot is the best of the lot because it can consistently watch for certain phrases or tags and immediately let me know when they have been referenced. It's not a community, but Ginabot doesn't sleep and she's always handy.
I've seen a few social media platforms coming go in my day as a side effect of being a tremendous geek and predating the web on the Internet. (I lament the loss of Usenet every day.) There have been enough come and gone that one more really won't make much of a difference to me.
If it does come down to it, however, a postmortem is going to dissect why an obsession with global metrics instead of local/individual ones is an overall mechanical failure.
@wizardave sometimes likes to crunch numbers. He did take a look at just one to find this out
https://steemit.com/steemit/@wizardave/are-people-really-getting-rich-using-bots-and-upvote-services
maybe you could show him a little ORGANIC love if it helps
@entrepreneur916 dba @thundercurator has a good handle on the numbers.
Sorry tarazkp you lost me.
Not all money which is sent to the bots does come from the reward pool. Lots of the SBD are bought on the exchanges, just to feed to the bots. At least that's what I do understand!
the money sent to the bots doesn't come from the pool, what they vote comes out of the pool though. The liquid payments buy access to the pool value.
If I do take a look at steem.supply it states that the reward pool/fund is:
Total rewards fund: 1,625,609 USD / 725,281 STEEM
You are talking about a reward pool of 90K.
Or does steem.supply refer to the weekly total of Steem?
https://steemdb.com/api/rewards
Here are the daily amounts paid out.
I wish I can help you with the math calculation @tarazkp. Too bad I may be worse than you 🙈🙈
Thanks for coming around my blog.👍
No worries, glad to help a little.
This is exactly the case, I am afraid.
There is just not enough incentivation to curate. It is actually sad.
The vast majority of steemians stay with a couple of authors, and read them on a daily basis. Just them.
It is real hard to get an audience if you do not have enough money to buy or invest here.
This is what I posted the other day. There are very few people looking around the site for interesting articles. I found you the other day and enjoyed your posts. I posted something on bot's the other day and it was more of a test as generally I wasn't expecting it to be read. Quite sad but true. It's a bit like fishing you cast it out and just wait. I re checked after about 45 minutes and no one had even opened it. I answered my own piece and replied what I already knew. If you are not established on the site already you are basically screwed. Growth potential is rather grim. Majority of users are using bots and they are not interested in what anyone else is writing. They are here trying to make money.