Mind Your Votes! An investigation and guide to maximizing your Curation Rewards.
As a Reddit veteran, I take my upvotes very seriously. Every upvote matters. When I first joined Steemit, I went about it the same way I have trained myself over the years. I'm scrupulous with my upvoting habits, only upvoting posts that truly intrigued me.
There's much more to Steemit than Reddit though. In this post, we will specifically discuss Curation Rewards. This was bit of a mystery to me. For one, the Steem Whitepaper contains precious little information about the subject, while the information on the website is outdated as there were wide sweeping changes to the algorithm just before the first payouts.
What really helped me dive deep into this subject was @jesta's brilliant SteemStats. Specifically, Vote Inspector, which breaks down the weightage of curations rewards by post. When it first released, I posted about my initial discoveries. I was surprised to find many steemers interested, so over the last 4 days, I have been investigating vote distribution patterns for several posts. I have also held several voting experiments myself, as well as followed those of some steemers with curious voting habits.
Here's what I found.
There are four things that determine your curation reward, in order of magnitude -
1. Eventual popularity of the post (Total SBD generated)
2. Your Steem Power (Your stake in the Steem pie)
3. Timing of your vote (How soon after the post was made)
4. Your Voting Power (How often you upvote)
Popularity - SBD Generated
Every post generates a certain amount of SBD. 25% of this amount is distributed among all curators. Note that unlike author rewards, curation rewards are exclusively paid out in Steem Power. So the total curation rewards for a post is 25% of the Steem Power equivalent of the total SBD generated. Your share of this reward is determined by the metric Weight.
For example, if a post earns $3,000, at an exchange rate of 1 SP = 3 SBD, a total of 250 SP are paid out to all upvoters. If the proportion of your Weight as a sum of all upvoters is 1%, you get 2.5 SP. Note that there's a minimum payout threshold of 0.001 SP.
Steem Power
In the curation world, Steem Power is king. Or, more accurately, your vesting interest in Steemit. I routinely see posts where ~80% of the curation rewards go to one user. Of course, these are mostly smaller posts. But even Trending posts can be highly skewed.
This popular post with over 100 votes has ~65% curation rewards heading to a single steemer.
Timing
Like kings, Steem Power is fallible. There's one way to beat highly vested steemers at their own game - timing. And, well, hoping that they are late.
The earlier you recognize the potential of a post, the greater your Weight.
Here, @bravenewcoin gets a larger share than @ned despite @ned have 700 times (!!!) the Steem Power! Similarly, @thedashguy beats them all. This is simply because @thedashguy and @bravenewcoin discovered the post sooner, while the whales came in an upvoted it 5 hours later.
However, there's a massive caveat -
The 30 minute rule
At 0 seconds after the post is made, you are penalized 100% of your reward. ( This decreases linearly till 30 minutes after the post, where the penalty is 0. The penalty fee goes directly to the author. At this moment, the penalty is linear. I.e. 100% at 0 seconcds, 50% penalty at 15 minutes, 0% penalty at 30 minutes.
This leads to an interesting dilemma. By waiting 30 minutes, you avoid any penalties. However, the post could explode with upvotes within these 30 minutes, and you are left with a negligible share at the end of it, penalties or not.
In the end, it comes down to a simple gamble - Do you think this is potentially a very popular post and could go viral any minute? Take the penalty. Be early.
This is one of my most successful posts - a big reward for a minnow like myself. I could see this post would be a massive hit and be trending in short order, so I took the risk. I voted in the first 7 minutes, received a 75% penalty. Yet, it turned out to be the right decision as minutes later, the whales joined in the fun. Within 30 minutes, the post had generated several thousand in SBD. If I had waited till then, I would have ended up with 0.001 SP, at which point any penalty is academic.
The other factor is time of day. Usually, during peak hours - Morning and evening US time, things trend very quickly. Often hitting $1000 within 10 minutes. Whereas off-peak hours, posts often take 30 minutes to a couple of hours to be picked up by influential steemers.
Certain posts are slow burners. The more analytical, longer and verbose posts often take time to catch on. Whereas some posts are punchy and trend fast - particularly visual ones.
My general recommendation is - wait 10 minutes for posts that are very interesting and likely to be popular. 20 minutes otherwise. If it's peak hour and a post is destined to be Top 3 on the Trending page, take the penalty. Vote within 3-7 minutes.
You are only rewarded for upvoting within 12 hours. Upvotes beyond that count for a long-term 30 day payout, but aren't rewarded for curation.
Voting Power
Finally, all steemers have a Voting Power, that is polled at the time of the upvote. The more frequent your upvotes, the lower your voting power. If you are spamming the upvote button constantly, your Voting Power is likely to be under 50%. If you rarely vote - less than 5 times a day - your Voting Power will be over 90%. This means each of your vote counts for much more.
In general, I have found a majority of steemers fall in the 70%-80% range, so this is mostly a non-factor. Whales usually have 90+% Voting Power as they are much more discriminate with their upvotes.
Generally, I have found 30-40 upvotes per day is the optimum, maintaining your Voting Power in the ~70% range. Between getting the maximum upvotes in without diluting each upvote.
Voting Power regenerates steadily to 100% over 5 days.
Further observations
Should I upvote my own post?
Under every post, there's this little box that's checked by default - Upvote post. By clicking this, you forego any curation rewards for your post. On the other hand, this makes your post more visible.
My simple recommendation is - hold your upvote. Especially if you are a minnow. You can upvote your post at a later stage.
Un-voting
Note that by un-voting, you forego any curation rewards for the post, even if you upvote later.
So what type of content is popular on Steemit?
The truth is, this is always a bit of a lottery. I have seen many great posts head into the void, lost in obscurity. Conversely, I have seen simple one joke posts trending rapidly. However, I have observed at this time these posts usually do well -
- Posts by popular / familiar steemers.
- Tools / apps for Steemit.
- Original content - particularly some form of visual art. Paintings, photographs, videos.
- Travel blogs to exotic places.
- Extraordinary subject matters.
- Dramatic personal life experiences.
- Most of all, the topic du jour. It was makeup two weeks back, marijuana last week, etc.
Of course, I bet all of this changes once Steemit matures and there's a diverse range of curators.
Flags / Downvotes
Downvotes are tied to Steem Power and Voting Power just like Upvotes. However, I'm not clear on how this counts towards your curation rewards, if at all.
Upvoting Bots
We love our bots, here on Steemit. In recent days, there has been an spate of upvoting bots. About 30-40 of them which are upvoting certain posts willy-nilly. It appears to be targeted mostly at the #steemit tag and certain keywords. Don't let them bother you, as their Weight is near 0. They are too early, often taking on 98% penalty, their vote spamming means they have very low Voting Power, and finally, they have low Steem Power. It is a perfect storm of negligible.
A Personal Note
Personally, I intend to go back to my Reddit upvoting ways. I shall be upvoting original and interesting content as quickly as I can - get the authors my share of the curation reward. I know this will compromise my curation reward, but it's fine - for me it's more important to play my part in grooming the community, however insignificant my upvotes may be. Also, as a minnow, I don't have much to lose anyway.
I'm also concerned that this system might disincentivize upvoting niche content, which may never make the trending page. The ability to find a niche and like-minded people is paramount to social blogging platforms. It's why we post online - because our real life friends don't know much about cryptocurrency or jujitsu. We can only have those deep conversations online. I hope the Steemit developers have a solution for promoting original niche content. With this diversity, Steem is well place to flourish.
As a closing tease, I have a special post about Steemit lined up for Saturday which I have been working on for two weeks. Stay tuned!
Thanks for this informative writeup. Hopefully I've chosen just the right time to upvote it ;)
Re: self-upvoting, it's true that
but you do so in favor of the post author, which is you!
Came here to say this, but looks like you beat me to it!
I'm afraid this post is likely to be lost in the void now! So it wouldn't have mattered. But who knows, with the 30 day extension now, maybe it can be revived. It's pretty weird how my preliminary article with just a couple of charts trended while this analysis which took me 4 days to write is going to be lost!
You have raised an interesting point! In posts that I tried upvoting, my Weight showed up as 0- i.e. no part of the Curation Rewards. As to whether or how that goes to the author, I haven't been able to verify. Would be great if someone with inside knowledge of the code could chime in. :)
The instant self-upvote rewards do go to the author (you) under the 30 minute rule so there is no harm there. The only reason not to use it is if you think the post won't get votes and you don't want to waste the voting power.
@smooth i have seen many users over here constantly upvoting themselves on every comment, but personally i dont like this way , i prefer to give rewards to someone's else for his good content in post- not to upvote myself to get anything more, i prefer to make it by myself doing posts that are helpful to other users.
The self-voting feature we were discussing only applies to posts not comments.
Voting yourself routinely on comments is questionable since comments usually don't get that much in the way of votes and rewards, and each comment vote uses the same vote power as a post vote. I only do it when I feel my comment is particularly important and I want to promote it, but in that case it isn't for the rewards, it is for visibility.
However, there are some users (I'm one of them) who don't post very much but are active on comments. They may feel that their comments are useful and a valuable contribution to the discussion, so worthy of rewards. (I'll leave it to others to decide if my comments are valuable.)
Thank you @smooth. That does make sense. Wish I could edit the post - this was clearly an oversight on my part, but it's already been locked! I just hope readers see this chain.
^^^ THIS
👍nice information, thanks @liberosist
@nevid. Thank you, this seems to be the consistent trend
Badda-bing, right there baby.
Ditto
Μy humble opinion is that it must be changed! It is fair that the curators get rewarded for their finds after 12 hours as well !!! Why not? The system right now don't give motivation for curators to extract quality content after the 12 hour span... This post was an exception! It must get the rule that good content will eventually survive !!! ...good content must survive regardless of the heroes out-there....
i absolutely agree. This post is as useful today as it was a year ago
Minnow here. Heck, I'm reading this now and I still find it as useful as ever.
I agree with you there. My guess is the current algorithm doesn't scale over such a long period, which would mean people voting after 3 days, let alone 3 weeks, get nothing. So there would have to be a separate curation reward algorithm for the extended period. Perhaps they are working on it and it'll come with the next Hard Fork. Wish @theoretical would step in with some insights.
There is too much incentive to game the voting system, and not enough incentive to cast genuine votes. Not only that, but I think people should be rewarded less when voting for authors with a good track record.
Right now curation on Steemit is like horse racing, except the favourites are given the same long odds as the outsiders. This is extremely polarising.
I designed the curation rewards system. The weight formula is tuned so that the initial value of your reward is about the same, regardless of which post you upvote [1]. This means that naturally your weight will be less if you're the last person to upvote. You can read more about the curation rewards math on my blog.
Thank you for chiming in, @theoretical! I'll certainly check up your blog, and I hope others would too.
If there are some Newbie like me, i recomend to read the @theoretical blog, there is a lot of info about the reward system. thanks a lot for all that info.
I keep learning more and more little quirks about how this works! Great read on how to maximize curation rewards! Thanks @liberosist! (cc @ftlian @dollarvigilante)
Glad it helped you, Randy. I know you are in a good position to pass on this knowledge on podcasts and people from the Free State Project :)
You know I thought that during 0-15 min all curation rewards go 100% to author and than between 15-30 it decreases until hits 0%.
that is why I tend to wait to 15 min mark for voting.
It's a linear curve. It used to be a power curve, but as of now it's linear - 100% at 0, 50% at 15, 25% at 22.5, 0% at 30, and so on.
that's good to know, thanks.
Great job organizing this content into a comprehensive guide. Very useful!
The steemstats.com doesn't seem to show anything yet for the 4 week payment. All the weights for posts after the initial payout are 0. Is there a reason for this? Do you know if it is something they are planning to update?
Like I mentioned in the post, there's no Curation Rewards for the extended 4 week period. All curation rewards are handed out in the initial 12 hour+ period.
Ok, that makes sense. So the "Vote Distribution" section is showing the curation rewards? I have been thinking it was how the votes/weights contributed to the payout amount. I assume they are similar/related, but not exactly the same thing.
Yes, it's the Vote Distribution section. And yes, the percent of Weight contributes to your curation reward. The calculation is pretty accurate too. To find out your curation reward, you can use this formula -
(Weight percent / 100) * 0.25 * Total SBD generated by the post.
Cool, thanks!
Finally the detail I been looking for - minimum payout of 0.001 SP. Plus, there was a bunch of interesting stuff in this post! Thanks.
I missed out on a little detail, too late to edit the post now!
To hit the 0.001 SP threshold consistently, you need a Steem Power of about 250. If your timing is right, you'll sometimes get curation rewards with 100 SP in the Wallet, but to consistently be rewarded, make sure it's well above that.
Unfortunately no rewards for flagging.(your voting power just decrease )
The good "flaggers " are heroes or big shareholders :)
Thanks for confirming. I suspected so - but this never showed up in the Vote distribution, so I wasn't sure.
I've seen @berniesanders spot a plagiarized post and wipe away a hundred upvotes in one go. :)
I don't see this post being lost into obscurity. It's the most informative and very well written. I love reading tutorials, especially about #steemit . Thank you for sharing this knowledge.
It WAS lost in obscurity! You'll see that it's 3 days old. It entered the void, like many of my posts do, with less than a dollar. Luckily, someone resurrected it with a big upvote. I think it was @berniesanders, the true Steemit superhero. :)
This is the new extended 4 week system in action - giving us a second chance. I hope some other great posts are rediscovered too, including some of my own.