The Impact of Unused SteemPower on the Rewards Pool - Blockchain Business Intelligence

in utopian-io •  7 years ago  (edited)

Almost two weeks ago I received a comment on a post from @freebornangel. The comment was lengthily and ended with

Yeah, thanks for making it even harder for the newbs to make a buck on their content

Ouch…I don’t want to hurt minnows…I’m still a minnow myself.

This was part of a response to the post https://steemit.com/steemit/@paulag/steemit-inc-and-misterdelegation-distributing-power-and-delegatee-ranking where I gave a shout out to @ned and @steemit to delegate more and not let their SP go unused.

I recommend you read the post and the comments left, but to summarize the discussion between @freebornangle and myself went a bit like this.

@Freebornangel

If the 75% of sp that doesn't vote did vote our votes would be worth even less than they are now.

Me: You are wrong!

@freebornangel: do the math and see

Me: Ok…I would love to prove you wrong.

The Aim of this Analysis

  1. To see what impact if any would a daily 100% vote from @steemit would have on my vote value
  2. To see what impact if any would 10 daily 100% votes from @steemit would have on my vote value
  3. To establish how much SP is unused

Calculating the value of a vote

I previously wrote an explanation of how to calculate the worth of a vote on steemit, which you can read here:
https://steemit.com/bisteemit/@paulag/how-to-calculate-the-worth-of-any-steemit-vote-steemit-business-intelligence

This posts covers the steps on where you can find all of the data along with the formula. To recap the formula is

a = (reward_balance/recent_claims)

b = (base/quote) // it is Steem/SBD Feed price

c = (total_vesting_fund_steem/total_vesting_shares)

d = steempower/c

e = power*weight // weight of vote for ex. 10000

f = (e+49)/50

G = f to inter

Vote amount in SBD = dG100ab

From this we can calculate the worth of a vote from @steemit

3.png

And we can also use the template to calculate the worth of my own vote

4.png

These values can be further verified at https://steemworld.org/@steemit

Calculations

So if a vote from @steemit is worth SBD $10,044.62 and they voted 1 time per day, what impact would this have?

Well there was a little math involved here

First I needed to convert SBD to vests. Using http://www.steemdollar.com/ 1 SBD = 1.0926 STEEM

And using http://www.steemdollar.com/vests.php 1.0926 STEEM = 2232.4950 vests

We can use these values to now calculate the Vests used per vote and multiple of votes for @steemit
And we also need to calculate the Rshares on a vote ( which I got the information from this post by @jfollas)

https://steemit.com/steemdev/@jfollas/write-a-steemit-web-app-part-13-how-to-calculate-a-vote-s-rshares

(there is probably a much easier way to work this out – I do tend to find the hard solution!)

The table below shows how much 1 vote, 10 votes and 90 days of 10 votes is worth in terms of SBD, STEEM and Vests. We can also see how much RShares are used at the same voting levels

4.png

Using this data, we can now amend the Vote worth calculation to include the change in a = (reward_balance/recent_claims). If we update this to include 1 vote from @steemit the change in my daily vote worth can be seen below

6.png

So with only 1 full 100% vote a day from @steemit would reduce my vote worth by 1.54%
If we update this to include 10 votes from @steemit the change in my daily vote worth can be seen below.

7.png

15% is a substantial and noticeable different on a vote value. However this is assuming that @steemit only started voting now.

In the original discussion with @freebornangle I made a counter claim that if the price of STEEM increases then we can offset this difference in vote value.

9.png

To offset 10 votes a day from @steem at 100% power the price of STEEM would need to increase to 3.94. That’s an 18% increase and very doable.

(EDIT - THIS SECTION IS UNDER QUERY AS THERE IS A POSSIBLE ERROR IN MY CALCULATION)
However this does not address the issue of the rewards pool. At the end of the day, the reward pool is only filled by so much. If the payouts are greater than the refill then the pool will eventually dry out.

So if we backdate the data, so that the reward balance and recent_claims include 10 votes a day at 100% for 90 days look what happens to my vote worth…

3.png

@freebornangel you were so right…..I take back what I said.

90 days of voting 10 votes a day at 100% weighted Power would have drained 9,825,747.73. Sure the rewards pools doesn’t have that much…lol….we’d be fucked!

All the data for above was collected 26 Feb at 22:30 GMT

How much SP has gone unused over 90 days

To find out how much SP was unused over 90 days I connected to the Steemsql database using Power BI

The query used was

 select *
 from Accounts NOLOCK

The data was collected on the 24th February and DAX was then used for analysis and calculations.
The table below shows data for ALL accounts on STEEMIT

11.png

The table below shows data for all accounts that have NOT voted in the last 90 days

12.png

39% of Total SP has gone unused in the last 90 days. Of this @steemit owns 60%. I’m not going to run all the figures again to adjust for the additional 40%, but if all the SP was used to vote 10 votes a day at 100% weighted power, at the current rate of refill then Steemit wouldn’t have much hope.

So my original shot out to @ned to delegate more power….well I retract that statement now

This does however leave me with some things to work out in my head.

How can we ensure that the rewards pool is not drained? And I’m not talking about spam and plagiarism, but what if we grow so big that there are many many good posts and many good curators finding these posts, could we be paying out faster than refilling? And if this scenario is possible, what can be done to improve the rate the rewards pool is filled at? Would love to hear your thoughts and comments on this.

Conclusion

@freebornangel is right (which makes me wrong) and I am delighted to now have a more solid understanding of the STEEM blockchain workings. This analysis also highlights the importance of the ‘Reward Pool’ and work done by those to ensure Steemit is free from spam and people are not raping the rewards pool.

Steemit needs ‘dead fish’ and unused SP. Imagine if all the STEEM on the exchanges was sent back into steemit and used for voting… I wonder how that would affect the price? – If there is no STEEM on the market, would people be willing to pay more even though the vote would be worth very little ? How can we find a happy balance?

I love when math proves me wrong, but I have concerns now about the future of the rewards pool and its ability to sustain massive growth on the platform.

I am part of a Blockchain Business Intelligence community. We all post under the tag #blockchainbi. If you have an analysis you would like carried out on Steemit data, please do contact me or any of the #blockchainbi team and we will do our best to help you...

You can find #blockchainbi on discord https://discordapp.com/invite/JN7Yv7j



Posted on Utopian.io - Rewarding Open Source Contributors

Authors get paid when people like you upvote their post.
If you enjoyed what you read here, create your account today and start earning FREE STEEM!
Sort Order:  
Loading...

Hmm... This is something I don't often think about. I suppose the reasoning is that as more users upvote demand for Steem will increase exponentially, bolstering the price exponentially. Your calculations rely on Steem increasing in price either not at all or substantially slower than it fries in popularity.

can the price of steem increase faster than the popularity of steemit.... I dont know but I do hope so

price of steem should be a prediction of steem's potential, so theoretically it should increase faster...

I see that the way to succeed here is to fulfil the original intent set out by the white paper, to ensure that the community subjectively rewards the best posts (as determined by the community and not just by one person or a small group). As long as posts reach the top due to buying votes (aka having the biggest wallet to begin with), people will reject the system and will view it as a kind of pyramid scheme. On the other hand, if posts are reaching the top due to their quality, then the system works and people will invest in the system for it's long term functional value, rather than for it's short term 'ROI'.

Increased investment means increased token value, which means increased value of rewards, which means more posts can be supported with a decent payout rate. The only way to make the distribution of slices of a cake more fair is to ensure that those cutting and distributing the slices are numerous enough that they can deal with the precision involved in a manageable way. Having 10 people voting with massive SP (delegated or not) can never be as effective as having 10,000 people voting with smaller portions of the SP.

I really want to note your integrity, on full display in this post.

"...@freebornangel you were so right…..I take back what I said."

I am deeply moved. I kid you not. I admire honesty, and honesty is impossible for many people that seem to find any opportunity to learn something they didn't know a personal slight instead. Without the ability to admit we were wrong, we can never become right.

I did not realize that were all the SP extant to be actively voting the rewards pool would completely drain, and that changes my views on things. Thanks for that too!

Kudos!

I like people that challenge the way I think and point me in the right direction....
Also there might be an error in my calculation above, I have added an edit as it is under query at the moment

Can you look into how the price rising makes it harder to build sp, too?

Most of the sp i have i got when the price was around .15usd.

My napkin tells me that were the price to go to 20usd getting to 500sp and the slider will be near impossible.

Slider is just a feature of Steemit. You can do the same on Steem blockchain with other tools. Busy.org is an example. Also as the price increases Steemit can reduce the limit to may be say 100 SP.

Did you see that bs that lexenbrise tried to rebut me with?
He did a good enough job that the easily misled would buy into it, but jebus, that was a load of crap.

Totally agree with @valued-customer. This is what drives great discussions and learning. Every post that @paulag presents I learn from ... That often includes the great comments. Willingness to change one's mind, given the data, is the mark of a good scientist.

Great post @paulag, I was actually contemplating some of these quandaries as I delegated some SP around. Timely post... Is there a global conciseness ;-)

thank you @jasonbu

There are 3 billion people connected to the Internet and that number is expected to double in the next 3 years.
Steem doesn't have a supply problem, it has a marketing problem.

Most certainly with more users, the price of steem has to go up. So in a sense it will level off if my vote is worth 1$ at full power so when Steem is 30$, my vote is 10$. As the value goes up more of that Steem can come into the pool and dilute it making our votes worth a bit less, and stalling the value from rising maybe.. I would like some sort of solution for rewarding good content without having to buy the votes since everyone with Steem power is delegating it to vote bots. its suck an intricate ecosystems its hard to see the unintended consequences of making changes to anything. Changes need to be made though

some sort of changes are needed, but I have yet to find a perfect solution...

Im still on a 100mv soft cap.
Not an enforced rule but a gentleperson's agreement.
And downvotes for gangs and vote sellers.

One way is to somehow elevate the status of promoted posts again. People are prepared to pay for visibility (and ROI) via the upvote bots. We can at least take care of some of the visibility issues by including promoted posts in our regular feed. Why's that exactly? well, promoted posts burn steem in order to be promoted. That has a small effect on raising the price of steem for everyone while at the same time giving the poster the visibility they want. For sure it doesn't give ROI in the sense an upvote bot does but it reduces the "i need visibility and will pay for it" argument in favour of the upvote bots.
Would you accept a promoted post or two, clearly marked as such, in your feed?

How did you come up with those data of unused SP? Steemsql?

yes, I got all the accounts and totted the controlling sp (owned, less delegated + delegation rec.) and then I filtered the data for accounts that have not voted in the last 90 days

Seems to me like you're asking the wrong question with "How can we ensure that the rewards pool is not drained?". Sorry, this might be semantics, but the purpose of the rewards pool is to be "drained" in the first place. By that I mean that the entire pool of rewards is meant to be distributed, with the only difference being the proportion of rewards going to any user.

The question should instead be: "How can we ensure that we maintain the (fiat) monetary value of any single good post over time, despite more users joining and posting good posts?"

Solution 1: I read your very informative post on calculating vote value, and if we increase reward_balance with time, we can help to maintain the vote value. The white paper stipulates a yearly inflation rate of 9.5%, so that will help slightly, as long as the increase in supply is met by an increase in demand.

Solution 2: Except the above, it seems like the only solution is for the value of Steem to increase. As long as the monetary value of Steem increases to counteract the fall in my share of the rewards pool, the overall value to me can remain unchanged. There is no other solution unless we change the formula through which payouts to each user are calculated. At the current rate of increase in users, the value of Steem should increase as well, because the increase in supply of reward_balance is smaller than the number of users.

The problem for new users, though, is that reward payouts are not calculated after 3 decimal places, and any payout smaller than 0.001 is not paid out. This is a problem that actual new users face. That is something I don't understand at all - it is easy to increase the precision of payouts, it'll just at a small cost of making the numbers a bit harder to interpret. If we rely only on an increase in Steem value to maintain the monetary value of contributions to the platform, then this problem will only worsen and become a natural upper limit on how many new users join and stay on the platform.

The reward pool is the inflation rate.

  ·  7 years ago (edited)

Do you mean instead that the growth rate of the reward pool is the inflation rate?

None of the rewards are given away from someone who has them. They are, in a sense, mined. Every reward in the reward pool increases the amount of steem (and/or SBD) that are out there. Rewards are inflation.

The rshares voted are paid, they only round down in the interface, is what i was told.
So even those votes that dont give enough vests to equal .001 still payout.

I'm pretty sure you are misunderstanding the way the rewards pool works here, but alas, do not hold the knowledge to make an argument. Perhaps @biophil can confirm your suppositions or my suspicions.

you might be correct here.....there is an error in my calculation...
In the calculation on 90 days past, I have deducted 90 days of 10 100% upvotes from the rewards pool, which puts the value as negative, but shown as 0 above. However I did not adjust for the 15% decrease in all other payouts as a result of the higher rshare on a daily basis. I am just trying to confirm this now and will make an update

Hi Paula

Good to have this cleared up.

Hopefully in time, the rewards pool will grow as the price of STEEM rises and inflation falls - I'm assuming this will allow Steemit to release more SP 'onto the market'.

It seems that Steemit inc know what they are doing :)

yes steemit inc know what they are doing.....but I dont think its as easy as they to release more sp onto the market

There is a yearly inflation rate that determines the size of the rewards pool.

  ·  7 years ago (edited)

This must be one of the main reasons SMT's are being pushed into production so fast and with such priority.

Redirecting user's attention to community driven tokens where crowd-funding can be facilitated and away from the current methods of delegating SP to big projects that have a much greater impact on the reward pool.

Nice one @paulag !

EDIT: Btw, the user you're referring to is @freebornangel, not @freebornangle ;)

thanks for spotting that - have updated.

I never considered SMT's in this and what you say makes a heap of sense.

Very interesting analysis, even if some of the calculations are being reviewed. Great work!

Thanks to @ecoinstant, this post was resteemed and highlighted in today's edition of The Daily Sneak.

Thank you for your efforts to create quality content!

After reading this article, I have 2 ideas:
You are definitely a math expert

  1. steemit idle sp, if you can play his role, the help of small fish is still great, because 99% of the whale's praise will not give the fish.

Wow, what an in depth and useful analysis. The whole voting structure is tough to comprehend and this is a great reference point from someone who's actually "done the maths". Conclusion is slightly worrying but as others have said I'm sure team steemit know what they're doing and new features like smt's will help ease the drain....

Wow, you handled that well. You stood your ground, did the math, and humbly accepted you were wrong. That was done with grace. Kudos to you, now you lost me half way through your math.

Help me understand how if content is created, voted on, that is draining the reward pool? What I really want to know is how do we increase the reward pool? I thought the more we voted and created content the more the pool grew not drained.

Thank you for the contribution. It has been approved.

You can contact us on Discord.
[utopian-moderator]

I like to read every post you...Upvote and follow you!

This is interesting. Observing what happens on Steemit which eerily exactly replicates what happens in the real physical world, you can draw conclusions about long term viability. Looks like playing short term game is less risky

There are a lot of other variables here that will ultimately dictate STEEM's price on the market, though. The "demand" side of the crypto market is not tied to "supply" as strictly as it is in more "traditional" markets, because so much depends on adoption and public perception. Either way, it'll certainly help to have at least some percentage (preferably a large one) burned or unused on a daily basis.

39% of Total SP has gone unused in the last 90 days. Of this @steemit owns 60%.

I was always intrigued to know these stats, so thank you for that.

The help people understand what you understood with your research I love to tell people that if they would be the only person to vote during a 30 days period their vote would be worth the entire reward pool which is currently close to $2.5M.

And to answer your question, if everyone would be voting that would mean there is a lot of value in voting and thus STEEM would be worth more. Their can never be 0 STEEM for sell because that would mean the demand is infinitely high and that's kind of an impossibility.

As for Steemit Inc delegation there are pros and cons. Steemit Inc has the most to lose from their actions. Hopefully they have calculated the risk carefully.

Gosh, @paulag, what a fascinating ponder! And seeing your math really made me happy that we have your brain on matters like this to help us see more clearly how this ecosystem is driven.

It is fascinating to explore how the different varieties/applications of STEEM/SP/SBD interplay to keep this current moving here.

Excellent exploration. Thanks for this and all you do!

Very interesting analysis, growth would be sustainable by the concurrent increase of the Steem price in USD, because basically the more Steemians register and use the platform, the more they want to buy Steem, and consequently rewards in Steem will decrease because of more competition, but rewards in USD will remain stable or even increase.

good work and well done for your humility :)

And thank you for introducing me to the #blockchainbi tag - it is highly relevant to my interests! Much easier to learn together than struggling on our own.

Hey @paulag I am @utopian-io. I have just upvoted you!

Achievements

  • WOW WOW WOW People loved what you did here. GREAT JOB!
  • Seems like you contribute quite often. AMAZING!

Community-Driven Witness!

I am the first and only Steem Community-Driven Witness. Participate on Discord. Lets GROW TOGETHER!

mooncryption-utopian-witness-gif

Up-vote this comment to grow my power and help Open Source contributions like this one. Want to chat? Join me on Discord https://discord.gg/Pc8HG9x

  ·  7 years ago (edited)

Doesn't it depend on how you look at it? Suppose they were upvoting you 10 times a day...

You might look at it differently. The reward pool is split (something like 75% to authors and curators, 15% to witnesses and 10% to SP holders). As far as this discussion goes the last two have no influence. Suppose they were only curating, that would indeed mean they would influence the 75% because of curation. That 75% is again split in 75% for authors and 25% for curators. Because of this our vote would be worth less, our curation 'income' would be less. But at the same time if their curation would be uniformly split (ideal world), wouldn't our author rewards be bigger because they would be voting on our content? Wouldn't this also cause the whale circle jerking to be less effective because their vote would be worth less and their 'author' reward would be less?

I'm not sure if this is correct. One drawback for us would be they would receive curation reward as well. The system and the implications are just so complicated it makes me doubt they did this by design. It's just one big experiment.

One thing I do note in the tone of the discussion is the payout reward. We must remember we are payed out in steem. More users would mean a more uniformal spread of the payout, and thus on average a lower steem payout. If you want to keep the same USD payout that can only happen if steem to USD goes up.

It’s nice to see numbers to what I have assumed for a while now. Increased price and less people using their full voting power are only ways.

I don’t think reward pool will increase at a fast enough rate. Even more so if bigger and bigger follows come over here from places like YouTube, Instagram and whatever else. I’m sure a year from now we all will have been grateful at how easy it was just to earn a single Steem. It will be interesting to see at what point do they increase the visible decimal places in our wallets as a higher Steem price will warrant that.

Some of the bigger holding of SP seem to get this and don't vote to their max potential. Or they least delegate it off to try and get it to the causes they wish to support.

Somewhere between all the vote selling and delegation selling there will have to be a balance. If large invests truly want their holding to grow in value they would be better off looking at long term increase in price over short term of just selling their vote to highest bidder no matter what it is.

A caveat to this math occured to me, the votes by steemit are limited to their percentage of total votes?
Ergo, if steemit's votes totaled 88% of the total vote weight for the day then they would give 88% of the reward pool?
This may mitigate, though not eradicate, the impact.
The example i have used is if 1000sp are voted and i have 100sp voting for me, then i get 10% of the rewards, but when a whale comes and votes an additional 1000sp, total of 2000sp voting, i get only 5% of the rewards.

Ergo when a whale, or his delegate, votes everybody else loses money.