Thank you @arhag!
I switched to this version, because I don't agree with the conclusions by steemit regarding the payout data.

I went through the list for everyone at < -1 SP, and identified 52 active accounts known to me as individual user. Those have a total of -2100 SP.
Then I compared the activity of the winners of 104 to the big stakeholders (pages of transactions on steemd.com). It's obvious that they didn't do a fraction of the curation work of them. The only reason they get that much is their incredibly high stake, which wasn't really hard to achieve. Even if small curators invested thousands of dollars and powered up, they see only neglegible returns because of the unproportional bias.

I will keep 004 in the background to switch over if the majority decides so, but only under heavy protest.