Is Steemit really a Zero-Sum Game?
Some people on Steemit seem to be using tokenistic game theory arguments in an attempt to justify massive self-voting and vote trading practices with the idea that Steemit is a 'zero-sum game'.
Many of us self-vote to some extent, but I'm talking about 50% or more
The implication is that everybody is (or ought to be) trying to extract as much as they can from the rewards pool, and that self-voting and vote trading are acceptable and inevitable.
Strangely, some of the same people also seem to be of the opinion that flagging against them for such behaviour constitutes 'economic violence', and should therefore not be used to prevent them from extracting as much value as they can from the rewards pool. This is an amusingly incoherent argument, which if taken to its logical conclusion, lays the foundations for a prisoners dilemma which needn't exist.
What is a Zero-Sum Game?
In game theory, this is a situation where each participant's gain or loss is exactly balanced by the losses or gains of the other participants.
(paraphrased from wikipedia article)
Since the Steem rewards pool is added to at a pretty constant daily rate, at first glance it could seem to fall into the zero-sum category. Thinking more clearly though, we can see that the value of Steem varies over time, so the rewards pool's value is not fixed and unchanging, and that the behaviour of network participants affects it.
Network Effects and Non-Zero Sum Economics
A network effect is an effect that one user of a good or service has on the value of that product to others.
(paraphrased from wikipedia article)
When you up-vote worthy content, you are adding value to the platform because you are 'signposting', and making it easier for others to find content they are looking for. This is precisely why curation is economically rewarded. If the ease of finding good content improves, this likely has a positive impact on the Steem price which increases the value of the rewards pool.
If you only self-vote, or predominantly engage in vote trading, you are not adding value to the platform, but may in fact be adding to spam. Much like a weed is sometimes said to be a plant in the wrong place, I think spam could be thought of as content in the wrong place (ie. brought to undeserved attention).
Final Thoughts
If you are under the impression that Steemit/Steem is a zero-sum game, you may treat it as such, and be jointly responsible for its deterioration through a self-fulfilling prophesy. If however you take the approach of enlightened self-interest, you may find greater ultimate rewards by helping the community and encouraging a sustainable platform.
I do personally think that it would help to spread the enlightenment if curation rewards were increased, so that once again, there is a 50/50 split between authors and curators. This should create more favourable economic incentives and would reduce self-voting, vote-trading and spam.
Here's an interesting video about non-zero sum progress in more general terms.
I'd welcome any discussion about this, and my mind is totally open to convincing arguments against what I've written.
Excellent points.
"If the ease of finding good content improves, this likely has a positive impact on the Steem price which increases the value of the rewards pool."
Can you think of any way we could incentivize rewarding curation more without cutting author rewards by 1/3rd? We need some way to make it game-theoretically correct in the short-term to vote for quality content reasons not personal reward reasons.
I know this is a band-aid on a gaping wound, but perhaps curation rewards could also modify reputation upwards based on a formula? A small perk.
"If you only self-vote, or predominantly engage in vote trading, you are not adding value to the platform"
To be a pain as a devil's advocate here, does this not presume without proof that self-upvoted content and vote-traded content can not be quality? What if they are?
"enlightened self-interest"
A philosophy we would all do well to mimic.
It's quite hard to think of changes to the incentives structure that would certainly help the situation, and also be acceptable for investors, and not negatively affect the Steem price.
Yes, you're right of course, if the self-voter is voting on good stuff, there's no problem ;)
I may have already mentioned this (apologies, lot of comment threads going on), but what do you think about tying voting power more to reputation as @valued-customer suggests?
If reputation was more important, that could become an additional curation reward (Whether it be the standard rep or a second curation rep) without tapping the reward pool.
If the curve isn't too harsh, that may not be much of a disincentive from users buying in. I doubt many new Steemit users (a lot of which aren't totally crypto-savvy) are buying in their first week, and it doesn't take long to get to at least the 30s-50s.
I think if the standard reputation couldn't be affected by self-voting, then that might be a good idea, but unfortunately I think many existing reps have now been distorted by the self-voting problem itself.
It may not be popular, but actually if in the next hard-fork, the reputations were recalculated not to include self-voting, this may help. Of course though, it wouldn't do anything to solve the problems where reps had been artificially elevated though vote trading and sock puppets.
An additional curation rep is an interesting idea that I'd need to think about more, but it sounds intuitively like it might be good. It does have the disadvantage making thing seem more complex for the user, although I guess it could be hidden from the interface.
Indeed, I was thinking of curation rep as being hidden from the interface.
If the curve for curation was changed so that the more of it you did, the more rewarding it was over long periods of time, perhaps that would help too. Similar end achieved as adding a curation reputation.
Id be willing to entertain moving to 60:40 or 50:50 reward:curation as a reaction to too much self vote abuse hence spam.
Thanks! I think anything in that direction would be helpful.
There must be a author/curation split level at which self-voting actually becomes less profitable than voting randomly. That would not be the optimal level either, but some curation percentage slightly below that would mean that it was easier to earn Steem with a quick glance to evaluate quality, than by self-voting. I might try and consider the maths more thoroughly later.
Yes, I see what you mean. I should qualify my statement by saying:
If the curation percentage is high enough, then it would only be more profitable to vote on your own content if it could compete with the quality of other posts. For example, at a 99% curation reward level, the most significant factor in deciding what to vote for would be the quality of the content (instead of whether of not it was yours).
I'm certainly not saying that having such a high percentage would be optimal, just that such an equilibrium level must exist. Your point has clarified however that a mathematical deduction of this level isn't possible.
Since VP is weighted by the holdings of SP, the actual point at which self votes may become more valuable than votes from the community depends on the SP of the account.
~10,000 minnow votes are yet unequal in ability to impart reward than one whale vote.
For minnows, self votes are silly. Since people like to receive votes, they also like to reciprocate when they receive them. HF19 leaves minnows with ~10 votes they can cast daily before causing their VP to be unable to fully recharge. By self voting a minnow can receive those 10 votes.
However, by properly curating and casting those votes, accompanied by relevant comments, on the posts of others, the entire pool of ~100 votes those others possess becomes motivated to reciprocate with votes on the comment. Also, by voting other's posts, and commenting relevantly, those votes can attract followers who might upvote your posts daily.
Self votes don't even compare to the returns proper curation offers. This yet ignores the real value of curation, as the intercourse with those whose ideas you find interesting is worth more (at least to me) than financial rewards.
I haven't addressed pandering - in which minnows seek to gain votes from whales. It is a problem, as it deprives the community of content of higher quality panderers might generate were they to actually concern themselves with something other than rewards. It also focuses rewards into those accounts already possessing substantial holdings, and clogs the comments section with irrelevant fluff.
This makes no sense. Regardless of curation level, by self voting you increase your rewards. Since this seems to be the only consideration of self voters, they'd still do it. Upvotes on a post increase the author reward for the post, and that's their goal.
This imbalance is untenable. The white paper states the intention was to distribute ~90% of rewards to ~30% of accounts, but, as the chart shows ~99% of rewards inure to ~1% of accounts. The various selfvoting, botnet voting, vote timing collusion, circle jerking cliques, etc., are degrading the ability of the platform to grow, and thus causing Steem itself to not appreciate.
It is the weighting of VP by SP that causes all these problems, and additionally makes of Steem a security in the estimation of the SEC (at least according to my understanding of the regulations), which is a whole 'nother universe of problems I suspect none of us wants.
Under a hypothetical 100% curation / 0% author reward incentive, you will make most reward by voting on the posts which subsequently receive the most votes, whether these are your posts or not. So the optimal strategy is absolutely to vote on the best posts. A 99% curation incentive is little different. Of course, at 99% curation / 1% author reward, nobody would bother to post, but somewhere (I suspect much higher than 25%) is an optimal level.
You are pretty much right, except that a) financial rewards aren't the only issue that determines participation. It actually matters very little to me, personally. But, I am weird =p, and b) because financial rewards are so important to so many, characterizing the most upvoted posts as the 'best' is inaccurate, unless by best you only mean 'the best post to game for financial rewards'.
Again, I believe the real solution to the problem that rewards are being too concentrated doesn't involve the split between curation and creation rewards, but rather how VP is weighted by SP.
I have posted numerous times on the matter, and my views have so far been but reinforced as time passes. Full and in depth exposition of what I think is the best way for Steemit to move forward would create a wall of text here, and that wouldn't be appropriate.
I'm not particularly against changing the split. I just don't think it addresses the central issue.
I agree. Jumping straight to 50/50 might be a bit too big of a change, but I'd be very curious to see how moves to 66:34 and 60:40 would pan out.
I think it would help reduce the problems and self-voters would be more inclined to spend their votes on quality posts instead of only themselves. So authors might end up getting a smaller slice, but of a much bigger pie.
I see what you're saying, but I suspect most self-voters are actually being rational agents, but with short-term preferences (possibly shorter than they intend!). If the economic incentives change so that some pretty minimal curation activity nets them more profit than voting on their own low-effort content, then the abuse will decline and/or their content improves to make it more profitable for them to self-vote on it. The network wins.
Do you think the abuse should be dealt with in some other way?
<"...you get 25% of what your vote was, the other 75% goes to the author, so if your vote is $0.12 then you get back $0.03 for voting."
This is much different that what I understood. I thought that curators received 25% of the post value, so that the curator in your example would receive $25.
I am now alarmed that it is my comments that produce over 2/3 of my rewards LOL
The problem with flagging is that there is a pretty severe opportunity cost.
A person flagging is granting rewards to all posts which have active votes instead of using their vote to gain rewards. Yes, some people will do it (perhaps gaining social capital) which is good, but if we simply consider rewards, and not reputations, I think this action actually isn't economically rational.
I think the 40/60 ratio. But it will not happen
I've stopped self-voting as I prefer to add value to the system at the expense of losing a little income for myself. Even if self-voting was banned people could still do it via other accounts they control. I believe that especially at this stage in the evolution of Steemit we need to be encouraging others and building the platform. If it goes viral and Steem goes to $100 then we'll all make much more anyway. It's about taking the long-term view and not trying to milk the system for all you can get. I appreciate that for some people a few dollars from here can change their lives, so they may abuse the system, but we can try to persuade them that another strategy could be better for them. It's all a big experiment and we have to see how it turns out
Thanks for that. I love that it is an experiment, that we all get to learn so much from it, and that it's not even the same thing to different people.
A well balanced article Andy on the zero-sum game when applied to Steemit :) UV/RS!
Thanks man!
You really do seem to keep a positive attitude ;-)>
Class always comes with a smile :)
True. An important point ;-)>
You Brits have always been good under pressure.
They really took notes during the Blitz!
I never self-vote, seems inauthentic. I rather give my votes to others so they can do the same.
I feel the same. I clearly value my own opinion, else I wouldn't share it with others. The point of this platform is for others to decide (and vote with their wallet) if they value my opinion as well.
Exactly!.
Good points. I understand the matter about getting visibility which makes sense. For those who invested their own money or are looking to make a buck, they should look at the long term implications of self-voting and if everyone else was doing that.
Truer words were never spoke.
To be fair, I totally understand that those who invested will not want a change in the incentives structure, but doesn't it come down to whether Steem is a short-term cash cow, or a long-term social media platform?
Maybe if the economic incentives changed, self-voting on comments for visibility could seem more reasonable because in doing so, you would be sacrificing some profit for visibility. That feels like a more natural trade-off.
The difficulty of analysis isn't as severe as the flag disincentive problem. I work on many such tools as @steemreports.
If somebody only has a profit motive, trying to change their behaviour though appeals to morality or long-termism won't work as well as changing the economic incentives.
You make a lot of really good points in this post. I also really love that there is so much conversation/debate on this post. That means it is a VERY valuable post to the community.
You are addressing a very important topic of which needs a lot of discussion.
I have not followed you before but I respect your opinion even more now that I see you have a decent amount of SP. Your speaking about self voting and investing not as a minnow but a legit investor. Lots of minnows or people not getting payouts like to complain but don't really have scope. You are an investor so understand how there needs to be a way for investors to profit of their investment in a simple and quick way. Right now its very hard to do unless you do self voting. So we need to re structure the system so that there are legit ways for big investors to profit on their investment without a lot of time.
Because most big investors do not have a lot of time to be on the site. Also most big investors will want to profit on their investment otherwise they will invest elsewhere.
In order to get the value of $teem to increase we need a lot more investors.
To get more investors we need more reliable and easy ways to profit of the investment.
Thanks for this important and comprehensive article and discussion!
SteemON!
Thanks man. I'm not sure whether some of what you say is tongue-in-cheek, but I'm very much a minnow. My only significant investment is my time, that said, it is significant (I develop @steemreports).
I know what you mean though, and I guess it comes down to whether Steem needs short-term investors or longer-term investors. I think to really prosper as a future massive social media platform, it needs longer-term investors who themselves promote the network effect through restrained self-voting. I do understand and have some sympathy for those who bought in on a shorter-term, more speculative basis though.
Yes long term investors are what we need, though for the most part the only way to get that is by providing them a reliable and easy way to profit. Otherwise they will invest somewhere else.
Because this is a social media platform and that is the value that backs the crypto it can be very complicated and ever changing based upon the masses. That is why this topic is so important to discuss.
Yes, it's a very interesting dynamic, and each HF changes it in unpredictable ways too.
Yes ur right...Another big factor....I mean once upon a time it was a 3 year power down....Now its just 3 months....Thats a HUGE change.....Life just always seems to make the rich richer. Rules, changes, decisions etc almost always made by the rich. Thats just as true here on Steemit as it is anywhere else.
Game theory is always working against those not playing the game...
I like that comment!
Congratulations! This post has been upvoted from the communal account, @minnowsupport, by Bless from the Minnow Support Project. It's a witness project run by aggroed, ausbitbank, teamsteem, theprophet0, someguy123, neoxian, followbtcnews/crimsonclad, and netuoso. The goal is to help Steemit grow by supporting Minnows and creating a social network. Please find us in the Peace, Abundance, and Liberty Network (PALnet) Discord Channel. It's a completely public and open space to all members of the Steemit community who voluntarily choose to be there.
I tried to tell some of the top people in steemspeak about this like 2 months ago, they didn't seem to get it.
The problem is the community is being forced to police what the network rules (coding) should naturally do in the most time efficient manner.
Time is the most valuable human asset and is THE thing humans are inherently trying to save/use more efficiently.
No other viral social media network in the world offers any curation rewards so I'm not sure why steem is trying to completely reinvent the wheel.
Its pretty clear High curation rewards ≠ best content with how centralized steem is, you'd have to add in another factor and or wait until its far more distributed.
It still seems like it will be too easy to game - the rewards are to great.
It naturally creates follower systems where everyone just joins trails to maximize rewards - so everyone follows X whales vote or X bots votes.
I cant find the post but I remember seeing something like "My AI bot is getting the highest ROI on curation rewards" follow my trail
Robot voice "Follow the bots the bots will think better then we can"
That is not how you get the absolute best content to the top of the site.
@ned @firepower @picokernel
That's very interesting, thanks for that feedback. It certainly does seem to make sense. I'm now following you.
I wasn't around when the curation percentage was higher previously - was this a big problem they were seeking to improve with the change to 25% curation?
Do you have any specific ideas on changes that you think may help?
Appreciate it :)
I don't think it's in mine nor the worlds best interest to disclose the systems publicly yet...
Ok. I look forward to hearing when the time is right then, or if at some time you would like some private evaluation of your ideas.
Systems theory is something I'm very interested in.
Someone please correct me if I'm wrong on this, but I think the zero sum argument depends on whether or not the system in question is closed. If new sources are being introduced at any time, zero-sum can no longer apply. So is the circulating supply of steem fixed or not? Anyone know for sure?
It's not. New Steem is created for the reward pool
I'm no game theory expert, but Steem is indeed an inflationary currency.
i think im beginning to get how the voting works,
tnx for this.
mhm, 50/50 sounds really fair.
Interesting post!
Reminds me of something one of my friends told me about his vision of hell - "everyone there all has giant chopsticks, but can't reach their own mouths." - If they all just fed each other and cooperated then no one would go hungry, but because they were selfish they were only thinking of themselves.
Thanks. That's a good analogy.
Welcome. Thank you.
i self vote all the time. just like now added 2p to my retirement fund.
Your outgoing self-vote percentage is actually only 23% :)
no idea what that means Andy haha it was only a joke (probably not a very good one) I am powering down.
Congratulations @andybets! You have completed some achievement on Steemit and have been rewarded with new badge(s) :
Click on any badge to view your own Board of Honor on SteemitBoard.
For more information about SteemitBoard, click here
If you no longer want to receive notifications, reply to this comment with the word
STOPMinnows need help
Well thx for all those interesting experiences... I'm actually really new on this platform but fully believe in it so im 100% power up all the time, its the only way to keep steemit strong and lasting... Upvoted ;)
Keep up the good work! =)
This post has received a 0.63 % upvote from @drotto thanks to: @banjo.
))))
Cool time pass
hope to visit it once :) have a great day