Looking in the Steemit reward structure bonnet
To many of us in the Steemit community, the reward structure is a bit of a black box. I would like to change that for myself and hopefully others.
What I know about the structure is that:
· Rewards are proportional to the number of upvotes you receive
· An upvote from a Whale can potentially be worth 100s of times the value of one from a newbie like myself
· The algorithm behind the reward structure is very complicated involving many different interacting elements
That’s about the extent of my knowledge.
I’m not sure about others in this community but I have always been interested in knowing how a reward for an article develops over the payment window and in particular, which upvotes contribute most to the total payout. To understand this you would need to record the cumulative reward for each additional upvote as well as the person who upvoted you and the time they upvoted. The data would look a bit like this:
With this data you could produce graphs like this:
To collect this data I plan to use a web scraper that works in the background and outputs the data on a regular basis for me to analyse.
Before I proceed with this I was hoping to get some feedback on whether there would be an appetite for something like this? The above example is just one way of utilizing this data.
It would be pretty cool to be able to put in your username and select one of your posts and the reward development stats for that article come up. This is a pretty heroic leap from what I plan to start with but if there was an interest I could expand this further.
Thank you for reading and I look forward to hearing your thoughts on this.
Yes that would be very helpful especially to us newbies who created some content then wonder why we received less than $0.50 while others are receiving hundreds of dollars with less number of upvotes. At least there will be more transparency...
The explanation is actually quite simple. Think of it this way:
1 STEEM = 1 Vote
So accounts that have more Steem Power have more votes.
The user interface is confusing because it shows how many accounts have voted on each post but not how many votes it has received (as defined above).
Exactly, I keep getting included in that 'most undervalued post' report. More transparency would be really helpful
Keep on rolling. We are behind you in this effort.
Popularity based rewards are all about whether a person's opinion has a particular weighting. This is obviously crucial but it does not necessarily promote quality and originality. Popularity can squeeze out true creativity as people strike a winning formula for persuasion.
To make the reward system more objective and allow those doing something new to make an impression, I think content analysis is a key thing for establishing the value of a post. For example, if I wrote an essay and embedded an old Youtube video then I am probably just using it to emphasize a point, or trying to squeeze that last bit of value from an old post to YouTube. But if it's a new video then there is a good chance it is being presented for the first time on SteemIt, or that it's a genuine new discovery on the part of the poster. I'd say those are worth more than older posts. Likewise a post which links to Youtube videos which have already been linked to in another post probably should have a lower intrinsic value.
I would have thought analysis based reward structures would be an interesting area to pursue in the fields of AI and machine learning.
Anyone else think that analysis of the content, especially the age, should have a considerably weighting towards the value of a piece. That way originality and variety can be rewarded as well as popularity. I think that would increase the quality of the content on SteemIt and prevent it from becoming the next MySpace which basically degenerated into a cluttered mess of self promotion.
Think I'm going to post that little reply as a stand alone article as I think it's an area to explore more....
Sounds interesting and worth the try, I would like to see what you come up with and if possible have a system of calculating potential payment....sounds good and all the best with it
Thanks, hopefully I'll have something over the next couple of days
This is a super helpful tool when you have it ready. I'm all for it! Thanks @bitgeek
I think value of upvote depends on the user who is upvoting. User with more steem power & value counts more then some newbie upvoting.
So when creating chart you will need to consider this as well for your analysis of upvoting.
I would be interested in seeing how the reward structure works. Just reading this helped me to see how to design my posts to be more productive. Thanks.
Just a quick question... I notice that some of my posts actually begin to decline in upvote value after a few days? Like I had one that was $7.50 and by the time I could claim it was $6.38. This happens with resteemed posts too. what is causing that. thanks
Thanks! The payout is just an estimate until the end of the Payment window so it can change depending on other factors. The total payout fund per day is fixed so maybe there were other articles after you that shifted the proportion awarded away from you. I'm not exactly sure but I think thats the reason. Same thing happened me!
Listening to this it looks like the rewards are connected to the market cap for the day... After some practice
Oh I see. My share is a percentage of something. that makes sense. thanks again
Interesting I am still learning all this stuff, still a newbie! Following you for more
Thanks!
This would definitely be interesting to learn more about.
Also, what do you use to scrape the website? I've been interested in developing a web scraper for a different purpose, but wasn't exactly sure where to start. Thanks.
Hi voidmu, I use Python but there are loads of options available. I'll hopefully have this ready in a few days. Thanks
This post has been ranked within the top 50 most undervalued posts in the second half of May 30. We estimate that this post is undervalued by $20.78 as compared to a scenario in which every voter had an equal say.
See the full rankings and details in The Daily Tribune: May 30 - Part II. You can also read about some of our methodology, data analysis and technical details in our initial post.
If you are the author and would prefer not to receive these comments, simply reply "Stop" to this comment.
Upvoted! :-D