Personally, I think one of the most useful metrics we could track would be, "payout per feed placement" and "payout per active feed placement".
- feed placements = unique (author's followers + followers of all resteemers + community subscribers)
- active feed placements is the same, but with some activity requirement (account active within the last 30 days/90 days, something like that)
If the blockchain is paying for attention, then feed placement count seems like a good place to see if it's "getting its money's worth".
As with any metric, I'm sure there would be some nuance, but if curators wanted to start regulating quality, I think lists of posts with high payouts and low feed placement counts would probably be a good place to start looking.
I keep thinking that one of these days, I'm going to write up some sort of reporting on these metrics and see what it turns up, but so far I haven't gotten to it.
Interesting to see @thoth.test showing up in your reporting. I'm the one who developed it. The account flips the curation model upside down, keeping the curation rewards from its own posts but sending all author rewards out to beneficiary accounts so that authors can continue receiving rewards from their posts after payout time and delegators can get passive rewards without a need for daily posting. I'm not sure why it sometimes receives small author rewards from its posts. Some sort of rounding gaps in the blockchain at payout time, I guess.
Thank you @remlaps. The idea of measuring payout per active feed placement is very interesting. It would approach the reward system from a completely different angle — looking not only at who receives rewards, but also at the potential audience and attention generated by those rewards. This could definitely be worth investigating in a future analysis.
Regarding @thoth.test, now that you mention it, I remember the project and how it works. I had simply forgotten about it when doing this analysis. That also explains very well why the account showed such a high number of self-comments and self-votes while the actual author rewards were extremely small.
Thanks for the reminder and for adding another interesting perspective to the discussion.