Hive has more actively voting stake than Steem

in #hivelast year

No doubt, the shutdown of SteemAuto had a big impact on the number of votes that are cast on Steem. @penguinpablo puts the number of votes on Steem and Hive per day into a graph, e.g. here, where we see a strong dropdown in the number of votes per day on Steem since around May 21st.

The number of votes alone is one part of the picture, but what about the stake behind?

I split all accounts on Steem an Hive into two groups and summed up all their effective stake (own + received delegations - outgoing delegations):

  • active: voted at least once in the week between May 24th - May 31th
  • passive: last vote was before May 24th

Here's the result:

ChainActive SP/HPPassive SP/HP
Steem55.69M SP107.39M SP
Hive99.09M HP46.93M HP

That's not even a close win! While Steem has a bit more total stake, Hive has almost double the amount of stake that took part in voting activities within the last week.

Looking into the number of accounts behind the votes gives the following picture:


This figure shows the total number of Steem and Hive accounts that voted at least once since May 24th. Also here, Hive is above Steem.

  • More active stake on Hive
  • More voting accounts on Hive
  • More votes on Hive

And with games and dapps on Hive, counting voting accounts does not capture all "active" users on the chain.


I was tempted to look at stake actively voting on Hive after seeing someone mention it in a comment the other day.

Almost double the SP/HP actively taking part is a huge difference. And the bots are back on Steem so i think the organic voting is way down, one look down my following feed confirms this.

I didn't expect the difference to be that big, TBH. Organic voting is hard to put into numbers, but I agree, there's not much left on Steem.

That, and a lot of the voting stake was locked or stolen.

yep, that makes a good part as well. The locked/stolen 23.6M stake is neither included in the active nor in the passive share since this was 'forcibly unstaked'. Not all of this would have been "active" with my criteria here, though...

Great one @crokkon!
I was curious about this as well.

Can you provide some info how did you get the active voting stake :)
A sum from the voters rshares or something else ...

Thanks @dalz, also for the reblog!
I did not go on rshares level here. I calculated the active stake as the sum of all effective accounts stakes (own vests + incoming delegations - outgoing delegations) for accounts that did at least one vote since May 24th (via last_vote_time), converted from vests to SP/HP

Got it ... thanks ...
Recently I tried that rshares thing and it was a bit to much data and I failed :)
The above looks more practical.

It would have been possible as well with rshares, but it's much more data. It would have weighted frequent voters stronger than casual/one-time voters, but calculating the result back to a reasonable metric like HP/SP is difficult.

Could it be because steemauto stopped working on Steem?

That certainly had a strong impact on the number of votes on Steem, roughly halving the number of votes per day on Steem. I can't tell from this data alone if the difference in the sum of voting stake has the same reason.

Pretty cool. I heard that some people are trying to develop something similar to Steemauto. 😀 But anyways Steemauto was open source and if anyone would really want to have that back again, they can take the code and setup their own version of it. I guess nobody did that yet. 😀

Yep, I'm pretty sure there will be a follow-up service on Steem. However, I'm not sure if it will be the same as before. SteemAuto was there for years now and I wouldn't be surprised if there were still a lot of accounts auto-voting even though the original owners were not active on Steem anymore. Those accounts would probably not sign up, even if there's a new autovoting service.

Ya that's a valid point. There were many accounts especially the inactive ones upvoting for years without any manual intervention. Now even if there is a new tool, people configuring and using it can be less in number.