I've been deeply interested about what witnesses do for the Steem blockchain, how their ranking is decided, and most specially about how they are getting rewarded. There are a number of articles written about witnesses but I find this one from pfunk most helpful for beginners, although it is old, and there's probably some information that needs updating, it still gives a powerful introduction on what witnesses are about.
This analysis will be dwell on the most rewarded witnesses of all time, and how that evolved to a time frame of recency. This would unavoidably touch on approval rating and activities that impacts the producer rewards.
The data points used in this analysis were acquired through the use of arcange's public SQL server database; Steem SQL and an excel spreadsheet.
Producer Rewards Distribution All-Time
Witnesses are paid in vesting shares for each blocks they produce. Being paid in Steem Power encourages further commitments from the witnesses to contribute through curation. The all-time rewards shown above does not really tell anyone what the value is in $. In the table below, I converted the vesting shares count into Steem Power, and took the $ equivalent using the average 7 days value of Steem at the time of the analysis.
| producer | Ranking in Top 50 | Reward in Vest | Reward in Steem | Reward in USD |
|---|---|---|---|---|
| xeldal | 14 | 1,639,327,203 | 798,812.01 | 1,357,980.42 |
| smooth.witness | 09 | 1,636,402,062 | 797,387 | 1,355,557.30 |
| abit | 31 | 1,624,265,781 | 791,473 | 1,345,503.89 |
| roadscape | 46 | 1,610,600,455 | 784,814 | 1,334,183.85 |
| arhag | No | 1,584,186,203 | 771,943 | 1,312,302.90 |
| riverhead | 18 | 1,583,504,727 | 771,611 | 1,311,738.38 |
| pharesim | 17 | 1,583,387,204 | 771,554 | 1,311,641.03 |
| witness.svk | 40 | 1,569,211,448 | 764,646 | 1,299,898.16 |
| clayop | 08 | 1,449,590,101 | 706,357 | 1,200,806.63 |
| complexring | 49 | 1,442,348,079 | 702,828 | 1,194,807.51 |
| blocktrades | 05 | 1,433,261,350 | 698,400 | 1,187,280.27 |
| wackou | 22 | 1,255,091,553 | 611,582 | 1,039,688.57 |
| steemed | 45 | 1,210,679,057 | 589,940 | 1,002,898.29 |
| steempty | 42 | 1,140,945,288 | 555,960 | 945,132.47 |
| silversteem | No | 1,046,777,161 | 510,074 | 867,125.79 |
| kushed | No | 958,475,733 | 467,046 | 793,978.94 |
| au1nethyb1 | No | 942,686,971 | 459,353 | 780,899.90 |
| datasecuritynode | No | 846,150,355 | 412,312 | 700,931.22 |
| nextgencrypto | No | 809,182,393 | 394,299 | 670,307.82 |
| dele-puppy | No | 804,616,245 | 392,074 | 666,525.33 |
| others | N/A | 9,836,128,188 | 4,792,952 | 8,148,019.18 |
I also added a column to say whether the producer is in the top 50 witnesses or not, and their ranking at the end of November if they are. That will make more sense in the next chart showing the rewards distribution in the whole month of November.
Note: At the time of the analysis, the database listed 14,130 producers. I believe this to include miners when Steem was still being mined. An explanation on how miners were rewarded was covered in pfunk's article linked in the beginning of this analysis
Producer Rewards Distribution All-Time
The most important contrast in the data when delimited to only the month of November, 2017 is that from 14,130 producers listed in the all-time database it went down to list only 113 witnesses who were able to produce at least one block, thereby rewarded in vesting shares. In the table below, I showed the same information as the one in the all-time reward distribution. In this one, the approval rating will make more sense.
| producer | Ranking in Top 50 | Reward in Vest | Reward in Steem | Reward in USD |
|---|---|---|---|---|
| riverhead | 18 | 16137532.43 | 7,863.50 | 8,885.76 |
| furion | 13 | 16135964.19 | 7,862.74 | 8,884.90 |
| clayop | 08 | 16135174 | 7,862.35 | 8,884.46 |
| roelandp | 04 | 16135172.5 | 7,862.35 | 8,884.46 |
| jesta | 01 | 16134782.47 | 7,862.16 | 8,884.24 |
| someguy123 | 07 | 16134403 | 7,861.98 | 8,884.04 |
| aggroed | 10 | 16134391.65 | 7,861.97 | 8,884.03 |
| anyx | 11 | 16133598.17 | 7,861.59 | 8,883.59 |
| xeldal | 14 | 16133213.75 | 7,861.40 | 8,883.38 |
| thecryptodrive | 12 | 16132820 | 7,861.21 | 8,883.16 |
| blocktrades | 05 | 16131631.76 | 7,860.63 | 8,882.51 |
| gtg | 03 | 16131627.13 | 7,860.63 | 8,882.51 |
| pfunk | 15 | 16130854.33 | 7,860.25 | 8,882.08 |
| timcliff | 02 | 16129682.45 | 7,859.68 | 8,881.44 |
| good-karma | 06 | 16126146.19 | 7,857.95 | 8,879.49 |
| ausbitbank | 16 | 16098285.1 | 7,844.38 | 8,864.15 |
| busy.witness | 19 | 14515717.3 | 7,073.22 | 7,992.74 |
| smooth.witness | 09 | 14380461.83 | 7,007.32 | 7,918.27 |
| pharesim | 17 | 14045031.94 | 6,843.87 | 7,733.57 |
| liondani | 21 | 10636673.14 | 5,183.04 | 5,856.84 |
| others | NA | 91231938.78 | 44,455.54 | 50,234.76 |
In this one you will notice that everyone in the top 20 most rewarded producers except one are in the top 19. Why is that? pfunk explained this in his article earlier linked at the beginning of this analysis. In his words:
The top 19 witness accounts with the most witness votes are delegated to produce a block every round. That leaves 1 block to be produced by a backup witness, and 1 block to be produced by a miner account every round. The backup witness is a proportional-to-approval timeshare of all witness accounts below the top 19 that have votes. This means that every account that has witness votes but is not in the top 19 have a probability to produce a block each round equal to their witness vote share versus the other witness accounts also not in the top 19.
roadscape's SteemD shows information on ranking, approval weight, blocks missed, and other information related to activities of the top 100 witness.
Some of of the information here explains why while blocktrades is top 5 in ranking in terms of approval rating, the witness is only top 11 in the most rewarded producer list in November. Among the top 10 in the approval rating, blocktrades had the most missed block.
Delimitation
This analysis is not meant to tackle profitability. This cannot capture the investments on equipment to run nodes, fund projects and development, and the spend on the effort to support the community.
Conclusion
Between the all-time rewards distribution and the November reward distribution, it will show significant difference in the top rewarded producer list.
- In the November data there are 17 producers in the top list who were not in the all-time top 20 most rewarded producer list. This tells us that the approval rating works. The ability to add or remove votes gives us the power to entrust the blockchain to only the most trusted members of the community in an on-going basis. The witness vote is of course Steem Power-weighted.
- In the all-time top 20 most rewarded producer, there are only four who stayed in the November top 20 most rewarded producer list. There were in fact 7 producers who are no longer in the top 50 of the witness approval rating list. This tells us that even the witnesses, just like content creators and curators also need to stay active and supportive of the community to get rewarded.
This analysis is just the start of my journey in understanding the witnesses in relation to Steem and Steemit. I hope that I was able to contribute to your own journey by doing this.
Posted on Utopian.io - Rewarding Open Source Contributors