Pimp my vote! Why your vote is working harder for you than it should be.

Imagine if everyone on the Steem blockchain stopped voting. Everyone except me, that is. In this dystopian universe you may think that post payouts would decline, that they would fall away to nothing; the trending page transformed into a barren wasteland, zero post payouts rolling across Steemit like tumbleweed.

But this is not so. Even if there were only one voter the total rewards distributed would remain broadly unchanged. It would be my votes, my votes of previously minuscule power, that would decide the attribution of everything.

And who would I vote for in this brave new world? Why, dear reader, I would vote for you.

So I’m gradually building up a model of the overall Steem blockchain, at least from a financial perspective. For the last week couple of weeks I looked at the market shares of the various platforms. Today I’ve moved on to considering reward allocation.

As we all know, post payout rewards are allocated in line with votes, with the size of these votes determined by the level of Steem Power a user holds: The more Steem Power, the larger the weight of the vote, the larger the allocation of rewards. So given this incentive, wouldn’t almost all Steem be held as Steem Power? And wouldn't everyone vote? These are the questions I attempt to answer below.

---

How much Steem is held as Steem Power?

Most users will be aware that not all Steem is Powered Up into Steem Power. There are large allocations of Steem that are thus not eligible for voting. These amounts can be seen in each users wallet, either held as immediately transferable tokens (STEEM which I will call Liquid Steem) or more securely in Steem Savings.

So let's start putting in some figures to build a picture of the breakdown of the Steem currency between these three categories. According to coinmarketcap.com there is approximately 250m circulating supply of Steem. By inspecting the steemsql database I can obtain a breakdown of the different holdings of Steem as follows:

	Steem Power	Liquid Steem	Steem in Savings	Total Steem
Little Fish	9,811,648	10,817,879	720,103	21,349,631
Minnows	8,301,197	40,848,354	225,575	49,375,126
Dolphins	19,104,950	1,469,060	54,766	20,628,776
Orcas	36,236,079	5,280,349	71,651	41,588,080
Whales	43,969,481	1,969,959	180,014	46,119,454
Steemit	71,296,692	4	3,372,916	74,669,612
TOTAL	188,720,047	60,385,607	4,625,025	253,730,679

Figures derived as at end of 29-11-2017

A few points to note:
- "Steemit" refers to the single user account @steemit.
- The marine classification is based on Steem Power, which means that investor accounts which hold no Steem Power, as well as the exchange accounts such as Bittrex and Poloniex, get classified as minnows, or Little Fish despite their huge holdings of Liquid Steem.

As can be seen, almost a quarter of the Steem currency is held as Liquid Steem. This is mainly on cryptocurrency exchanges: 39m is held by Bittrex and Poloniex alone. I assume that the reminder is in the accounts of investors who are not otherwise participating in the Steem environment. The majority of Steem Savings is held by the main Steemit Account.

---

Does everyone with Steem Power vote?

This is a tougher question. To investigate this I took a period of one week, from 23 - 29 November, and extracted the vote information by user from steemsql. Over this period 43,400 accounts voted.

I compared the amount of voting of each user to a theoretical full voting level (derived as 10 maximum votes per day x 7 days = 70 votes or 10 x 7 x 100% = 7000% as full voting). So for example, someone voting 5 full votes per day would get a rating of 50%. Someone voting 10 full votes per day would get a rating of 100%. Someone voting 15 full votes per day would get a rating of 150%.

I separated up votes from down votes and reapplied this process. I then capped each rating at a maximum of 200% (or -200%) to deal with over-voting. Finally I applied these ratings to the amount of Steem Power held by each user to determine the amount of Steem Power applied in voting over the week.

Here are the results:

	Steem Power	SP used to upvote - %	SP used to downvote - %
Little fish	9,811,648	13%	-0%
Minnows	8,301,197	54%	-2%
Dolphins	19,104,950	73%	-2%
Orcas	36,236,079	78%	-3%
Whales	43,969,481	43%	-3%
Steemit Account	71,296,692	0%	0%
TOTAL	188,720,047	35%	-1%

Figures derived as at end of 29-11-2017

Overall only around 35% of available Steem Power was actually used for upvoting in the week analysed. This equates to approximately 26% of total circulating Steem. Around 1% of available Steem Power was used to return rewards to the reward pool through downvoting.

Brief conclusions in understanding this 35% result:

• The Steemit account and a number of whales do not participate in reward allocation.
• I believe that there is a significant amount of SP tied up in very small "little fish" dead accounts.
• Minnows, dolphins, and orcas participate fairly robustly in reward allocation, with the involvement increasing at each higher level (this seems logical given the increased incentive per user)

As a net result, approximately 25% of circulating Steem determines the allocation of rewards. Each vote is therefore worth around 4 times its potential theoretical value if all Steem participated in reward allocation.

---

Questions

This has been a particularly unwieldy investigation, with a number of approximations necessary so please use the results with caution. If you have any questions or spot any errors please do not hesitate to leave a comment.

That's all for today. Thanks for reading!

---

Methodology and Tools for Analysis

Raw data was obtained through sql queries of the TxVotes, and Accounts table in steemsql (thank you as always @arcange!) using Valentina Studio.
Data was analysed in Numbers, the mac spreadsheet tool.
Data was obtained across the week of 23 -29 November 2017.

---

Posted on Utopian.io - Rewarding Open Source Contributors

---