new users vote much less than established ones. so during a userbase spike, the voting percentage would be lower. the shrinkage in absolute terms could be due to a lot more engagement in comments. if you are actively commenting maybe less likely to be curating? stats on comments would be interesting
A 58% drop in absolute voting seems pretty surprising. Are people shutting down tgeir curation bots? Are minnows getting discouraged that they can't earn curation rewards (due to rounding effects)?
The huge spike in voting was due to bots vote spamming. It was prevented with a client update by the witnesses hence the drop off.
Thanks @rainman I will make an EDIT
I think a lot of bots are gone :) I haven't seen a few bots which have vote/comment a lot.
Yeah, the reputation system cleared out all the commenting bots. Voting bots are alive and well, though. Just ask @jl777 - he has a bot that's voting with his main account.
Replies = Comments. Apologise for not making that clear. Also fell this week...
i figured as much, but how are you counting a reply/comment?
I would use a filter just based on length to get rid of most of the bot spam. counting all of them will contaminate the data and skew results. with the bot voting being neutered, we can see a big drop in bot spam, which would appear as much less comments unless you filter out the likely bot comments
Think you've got it there. My numbers are gross numbers off steemle.com
I have noticed a big fall in spam bots. I will add an EDIT in...