Mathematics for STEEMIT: the Pareto Principle, or how to make money in STEEM more
The Pareto principle is an empirical relationship, in simplified form, formulated as "20% of efforts give 80% of the result, and the remaining 80% of the effort - only 20% of result". As a rule, this principle is always observed in systems with complex relationships, in which the end result depends on the action of many factors. So, according to the Pareto principle of the many significant factors are just a few.
In life it helps to focus on really important to achieve the result of the action, discarding the unnecessary. But if we're talking about Incentive, then the Pareto principle can help to earn more. If you want to know how - read the conclusions at the end of the article, and a little math to prove my words.
STEEM is a decentralized system, which has already involved thousands of people. It would be logical to expect that in a system with so many ties the Pareto principle will be carried out. But, nevertheless, this empirical conclusion requires a rigorous mathematical verification.
For this I decided to investigate how to distribute the income of authors for posts. For analysis were taken from the incomes of authors as a top-level, i.e. without taking into account the earnings on the comments to the articles. Then for each author received his / her final earnings from publishing articles and received an array sorted in descending order of earnings of each author.
In total, the array came 9157 authors, earned a total of about $ 2.2 million. Dividing this amount by 10% intervals in $ 220,000 a schedule of the number of authors total income that falls within the interval from 10% to 100.
The diagram below shows the final schedule for convenience of analysis of numerical values are given in the table below the graph:
What is clear from these data? First, it is clear that the assumption of the principle of Pareto was fair. The Pareto principle is based on power dependence in logarithmic coordinates the power function looks like a video - what we see on the resulting graph.
In addition, the Incentive is uneven in the income of authors is much higher for the classical distribution is of the form we have a 80/20 proportion, when 90% of all income received only 6% of authors.
Practical insights
Confirming that the income distribution of authors for the publication of content is subject to the Pareto principle, you can make a practical conclusion - all the consequences of the Pareto law fair to the authors on the Incentive:
- First of all, know that most of your records will not get the reward. This is normal and not worry about it. Pure math.
- What you see is not always true. The Incentive there are many different factors affecting the earnings of authors. And most of these factors are not amenable to the identification and formalization.
- However, significant factors that affect the ranking of copyright content a bit.
- It is impossible to predict what content will work and what is not. From a practical point of view, you need to try different ideas and if some idea works - to continue to develop it until it works.
Another important consequence: there is a very small number (about 500) of all authors who earn 90% of the money on the Incentive for the publication of content. As we know, the authors get 75% of the remuneration per post, and 25% goes to the supervisors who voted for the lifting of the write-up. The practical conclusion is very simple - enough to vote for this small number of authors and is your reward for coaching will be the maximum
Moreover, such a vote is very easy to automate, since it is not necessary to carry out the semantic analysis of the text. Just saw a post of an author from the list - vote, will not lose. Some of them do. I found at least a few dozen bots, voting under such a scheme. But that's a topic for another article.