This post has been in the works since Hive began. Time and time again, I see that newbies are pretty confused by how it all works. So they should be, as the Hive curation system is over-complicated and scores bottom barrel points on user-friendliness.
So, I'm resurrecting an ancient series. Fun fact, Mind Your Votes! Parts I and II are two of my most popular posts. When the view counter was active, I once caught Part II going at 30,000 views. This is because, at one point, any combination of "curation" and "steem" would lead you to these posts. Even today, any combination of "curation", "steem" and "guide" or "help" etc will lead you to those posts.
Anyway, this post will be somewhat different, a refresher of how the curation rewards system works, and how you can do to maximize your rewards, aimed at newbies. It's not going to be a technical deep dive like some of the previous entries in this series. But more importantly, I'll consider this a Q&A post. If you have any questions about curation, ask below, and I'll answer.
Curation is a game of HP holding
The first tenet of curation rewards is - the greater your HP, the greater your rewards. It scales linearly, more or less. However, if you are a whale, you are likely to see diminishing returns as fewer HP holders larger than you will vote after you.
Currently, out of the pending payout, 50% of the rewards go to the post's creator, and 50% to curators. As a rule of thumb, you'll get at least 50% of the total contribution of your vote. Probably more, based on who votes after you.
Curation is a game of earliness and popularity
However, if you don't have significant HP holdings, you can still do better than you'd expect by playing the game of time and popularity.
The simple idea behind the curation rewards algorithm is that those who are early to popular posts get the highest rewards. There's a quadratic relationship, so the earlier you are, your rewards increase super-linearly.
Let's consider the ideal case - you'd want to vote on a post that's sitting at $0.0x, and then see it go to the top of trending. You can make pretty decent rewards. Due to the quadratic nature, you can even end up with situations where a very early 10K HP account ends up with greater CR than a late 1M HP account.
Being too early has penalties
However, if you vote before a post hits the 5 minute mark, your curation rewards are slashed and donated back to the reward pool. For example, if you vote on a post that's only 2.5 minutes old, and you're expecting a 10 HP curation reward, you'll end up with 5 HP. Tip: On the default Condenser/Hive.blog UX, the counter moves to "5 minutes old" when it's actually only 4 minutes and 30 seconds old. So, you might want to wait 30 seconds more to get all of your curation rewards.
Popular authors can have stiff competition
However, certain popular authors that are known to do well have bots and human curators alike queuing up to be the first to vote. This often leads to situations where an early voter will willingly sacrifice some of their curation rewards to be early. For example, voting at 4 minutes means you give up 20% of your curation rewards. However, if the popular author has a slew of votes incoming at 5 minutes, it's still well worth it. Remember, being early increases your curation rewards quadratically, so it can be well worth losing 20% to gain, let's say, 50%.
Seek undiscovered posts instead
If you're a manual curator, the game is rigged. Don't bother with popular authors, you'll never be able to compete with the bots. Instead, look out for quality content by less known authors, who don't have bots queued up for a mad dash in the first 5 minutes. It'll be much less headache for you, and you'll be doing what the system encourages - voting on undiscovered authors. Find posts which are setting with very little pending payout, vote on it, and then try to spread the word. Even if the post doesn't get much votes after you, you're still guaranteed the 50%.
If you're a bot, then yeah, just play the frontrunning game.
It's not all about posts with low pending payout
It's actually about predicting votes that come after you. So, it can be much more profitable to vote on a post that's sitting with a pending payout of $5, but ends up at $50; then a post that's pending payout of $0.00 and ends up at $5. Of course, if you are one of the last people to vote, you'll get relatively much fewer curation rewards.
Voting Power and voting strength
For all accounts above 500 HP on Hive.blog (or lower on PeakD), you get a voting strength slider, between 0% and 100%. This does exactly what it sounds like - you can modulate the strength of your vote. If your regular 100% vote gives out $0.10 in rewards, a 50% vote will give out $0.05. Of course, your curation rewards would be halved as well.
You get the option to modulate the strength of your vote because each account has a limited Voting Power, which you can check on Hiveblocks.com/@yourusername (or visible on PeakD). Each 100% vote you make decreases your VP by 2%. E.g. if your VP is 100%, a 100% strength vote will decrease your VP to 98%. If at 80%, it'll decrease to 78.4%. A 50% vote, however, will only take the VP down to 79.2%.
VP regenerates at a rate of 20% every 24 hours. So, the goal is to find an equilibrium between making as many votes as you can, while keeping the VP hovering less than 100%.
Always keep your VP below 100%
Any time spent with your VP at 100% is pure wastage. So, be sure to keep your VP under 100%. I aim for 80% or so, which gives me a leeway of 1 day without voting. You could go lower, of course, but the trade-off here is that'll reduce the maximum vote you can give. I.e. a 100% vote with VP at 100% is worth twice as much as a 100% vote with VP at 50%.
If you're unsure, follow good curators and curation projects
If you'd like to vote, but you're unsure, consider following the trails of curation projects and curators. Hive.vote is the best place to do so. Note that depending on where your vote lands, you can make a killing in curation rewards, or merely be on par. Because some of these trails have massive HP behind them, chances are this is probably the best way to maximize rewards.
Of course, if you have a bot that's fast, this is the way to maximize your curation rewards. As I've demonstrated in that post, you could easily make 20% APR by knowing the right tricks.
Don't bother, just vote on posts you like at the time you find them
There's more shenanigans involved, like the free downvotes, the final voting period etc, but I couldn't be bothered. If you think that this system is ridiculous, you'd be absolutely right. So, let me just say, whatever you read above is likely a waste of time for most normal humans. I feel like a bit of a dolt for wasting my writing this post, indeed.
Just find good content, preferably from lesser known authors, vote on them accordingly, and try to get the word out. Perhaps the only thing to keep in mind is to modulate your VP and keep it under 100%.
5 minutes Mark? I thought we were back to one minute?
I could very well be mistaken, but I think 1 minute was proposed but never made it to the hardfork. I'm trying to look for some documentation, but nothing definitive is available at first glance. Hope someone can clarify.
I simply vote whenever consumed something and feel like tipping, plus getting paid for my time as well, and even if I don't hit those marks where the reward is highest, I'm still earning more than I would anywhere else for consuming content, and you in this case earned more for creating it. So simple.
Exactly this!
Hmm, we've been here a while, and knew not to vote initially, not the other info though, thank you for the explanation.
Great info here. I do lament the fact that autovotes have gamed this system at the expense of true content discovery.
Yeah, it's a shame. While it can be tweaked for somewhat better results, ultimately an open platform means bots will likely always be dominant under the current paradigm. The only way is an overhaul of the curation system.
Interesting. Someone was saying: If you receive autoupvotes, manual curators will not vote you.
What can publishers do then?
I have just started receiving a small amount of auto-upvotes and I don't know how to feel about it:
That might be someone's personal opinion. I've seen plenty of manual curators vote on top of auto-upvotes, this post being an example.
There's not much you can do, just keep creating good content, engaging with others etc. Over time you'll build an audience.
Right. Indeed a personal opinion – just that it seems to match my personal experience compared to my days on Steemit.
Otherwise I am absolutely planning to do what you have recommended - create good content, engage, and build the audience.
Glad to see you are active! :)
This is one of the best article I have read so far.
Thank you for doing this!
As a content publisher, I really hope and wish a lot of curators can discover and value good content.
Going back to your earlier posts - and I see why we (publisher + curators) are in this together.