You are viewing a single comment's thread from:
RE: Steem Data Services (SDS) / Update Notice / Version 0.1.7b
Hi @steemchiller,
I was curious how you calculate the average efficiency and weighted average efficiency (and i guess by extension the efficiency) for the pending curation rewards in your dashboard. I was hoping to figure out how to store that metric for posts my bot has voted on after the rewards payout. (I think it would be more accurate since the number will be finalized).
Thanks so much for your help!
Hi @cmp2020,
for calculating the efficiency of a vote I compare its (outgoing) amount to the (incoming) curation reward. The sum of all collected efficiency values are dividided by the number of given votes to get the average.
The weighted average is calculated almost the same way as the normal average with the difference that the voted percentage is being taken into account.
For example, if one fails badly with one 1% vote but at the same time makes a very efficient 100% vote, the total weighted efficiency will be better than the normal one, which only looks at the number of votes and which always sees 1% as the same as the 100% vote.
So, you need to calculate an efficiency factor first (vote.percent / 10000) for each vote and multiply the efficiency and the count (1) by it.
I hope this helps a bit... :)
Thanks! @remlaps came up with a really creative way of calculating efficiency for paid out posts. He did the ratio of curator_vests to total_vests from the payout transaction over the ratio of curator_rshares to total_rshares in the post.
(curator_reward_vests / total_vests)
~-----------------------------------------------
(curator_rshares / total_rshares)
I thought this was a very smart way of calculating the efficiency without having to convert vests (since the conversion factor is always changing). I think in this case, 100% would represent the rewards the curator earned being equivalent to the amount of value they provided to the post, and more would be the curator earning more than they contributed, and less would be less.
I'm not sure if this efficiency is comparable to your efficiency, but I think it is a good metric for evaluating my ai's performance after payout.
Hi @steemchiller. I'm sorry to bother you.
As you know Steemit has a page https://steemit.com/promoted but it is not very popular. There are reasons for this, but one of the reasons is that it is not that easy to find. I wanted to suggest that if you like it, add a link to this page in your tool somewhere here:
Of course, I have no idea how difficult it is to do and whether it is even advisable, so the decision is yours. Thanks.