How many of the upvotes are done by paid voting bots?
I wanted to know how many of all upvotes on a daily basis are being done by paid voting services. So I dived into the Steem blockchain and counted all upvotes by all known paid voting bots.
I have made a list of all the known voting bots using https://steembottracker.com.
Here's the result...
Approximately 1.7% of all votes are done by a paid voting service. And as you can see in the chart, the paid votes are quite in line with the total number of votes.
Daily number of votes
| Date | Votes | Paid bot votes |
|---|---|---|
| 2018-03-13 | 745,468 | 11,803 |
| 2018-03-12 | 748,848 | 11,604 |
| 2018-03-11 | 711,500 | 11,283 |
| 2018-03-10 | 662,208 | 11,369 |
| 2018-03-09 | 640,175 | 11,726 |
| 2018-03-08 | 737,293 | 13,058 |
| 2018-03-07 | 777,449 | 13,716 |
2. Weekly number of votes
| Date | Votes | Paid bot votes |
|---|---|---|
| 201810 | 5,078,544 | 88,376 |
| 201809 | 4,935,492 | 89,606 |
| 201808 | 5,099,772 | 89,875 |
| 201807 | 5,083,608 | 86,787 |
| 201806 | 4,712,805 | 77,204 |
| 201805 | 4,931,440 | 86,818 |
| 201804 | 4,675,230 | 82,053 |
What portion of steem rewards (re)distibuted through voting bots?
Could you calc it?
I keep meaning to find the time to do that analysis. But it's pretty difficult as:
o There are many kinds of bots / purchased votes - those on steembottracker but also things like minnowbooster (which upvotes through different accounts so is harder to track, although possible) and "black market" vote sellers too.
o Working out the value of the votes rather than the number of votes is a heavier calculation if you want to do it accurately.
I'll get to it one day!
I think we as a community should come to understanding how bad is it.
Economy of steemit is of great importance and often neglected by developers of the project.
Users are actively using their SBD and Steem through bots making them the only crypto that is actually used in a functioning economy. Isn't that good for Steemit as a whole?
Steem indeed is the ONLY crypto where you 100% understand why you buy it.
Bots are the application where you could actually spend your crypto and enjoy the service they provide.
But is this healthy? I am not so sure.
If they are used to promote quality content they are a good thing in my opinion.
But I see the worst content upvoted for more than 50 SBD. I think the bot owners have a responsibility together to keep Steemit as clean as possible from these content farmers.
But if we can tackle that problem I think paid upvoting is a good thing and makes Steemit into a functional economy for quality content.
What I would find interesting is which group is using these bots more. Are it the already established Steemians or is everybody using them.
I do have a feeling that it would be mostly used by the more wintered Steemians and the usage will decrease when going down the foodchain.
Also requested this at Penguinpablo, if it is possible to dig this up.
Great work guys.
Exactly! I wondered about that too!
@bronevik Talking about bots and curation trail for minnows or anyone with low SP. I guess its really not helpful right? Please enlighten me
Votes less than 0.01 sbd are not even counted anywhere. Most minnow and low sp votes are just load for the blockchain.
Hmmmmm I see. So if I understand clearly, you're saying minnows need not bother voting people. I guess so
I've seen several, they say somewhere around 20% directly, but then another percentage is accounted for in bot follow votes to get curation money. Also, this article doesn't take into account the fact that the bots are using delegated votes, so that one bot casts many votes, under different names. I don't think there's any validity to this math at all.
I only discovered steembottracker.com the other day and it is interesting to see more information related to voting bots.
Could you please assist in providng more information on voting bots:
Hello @mkdouglas Someone said votes less than 0.01 sbd are not counted anywhere, that most minnows with low sp votes are just load for the blockchain. Please how true is it? Can you please explain better to me about it?
That's not true. Each vote is stored on the Steem blockchain.
I think every cent counts, at least up to 0.001 which is the smallest fraction that can be transacted on the Steem blockchain.
I also think all Steemians, even minnows, are useful hooks to catch more users. Network effects works on the fact that somebody you know is already using Steem.
The billions of Facebook users were attracted to Facebook by the first few million users who again were attracted by the first few thousand users. It grew from college students to middle aged parents to aged grandparents.
Wow so much insight. Thanks so much
Interesting. I am curious how many of those upvotes are self upvotes? I wrote article about necessary changes in Steemit in my opinion:
https://steemit.com/steem/@cicbar/steemit-necessary-changes
Please I'm curious about using bots and curation trail for minnows or anyone with low SP. I guess its really not helpful right? Please enlighten me @cicbar
Nice one - I think the Cash / crypto / steem reward generated via bots or booster services is more of interest - @felixxx did a research on that - roughly 16% - check his post:
https://steemit.com/steem/@felixxx/how-much-of-the-rewards-get-distributed-by-boosters
Self voting for people that might be interested in that, check out the post by @felixxx
wow thats alot! Thank you for linking. If he is not far from the exact number, 16% is quite worrisome for the long term health of the platform...
Nice post. It's helpful to see the data. While only ~2% of votes are by paid bots, I'd guess the percent of voting power from paid bots is much higher.
Not sure I love the multiple y-axes in each plot... it's easy to mistake to the votes and paid votes as being on the same scale. I'd recommend plots like these (from my recent blog post):
On the left you've got raw counts (in your case the two categories would be paid votes & unpaid votes). On the right you've got normalized votes. You'd probably want to zoom in the y-axis, given that such a low number of votes are paid.
An alternative would be a faceted plot, where the paid and total plots were vertically aligned but separate.
Keep up the great #datascience!
After a little more thought, I think the best visualization would be two vertically-aligned plots with the same x-axis of date. The top plot would be total number of votes per day. The bottom plot would be percent of votes by paid bots (y-axis zoomed to only the dynamic range). This would capture all the information above without any confusion.
Is the full daily/weekly data you created available?
I must say, I thought it was a whole lot more! It puts things in perspective. Off course some bots are huge whales, but there is really not much we can do about that...
As long as the big majority is normal traffic, I think the platform an keep growing!
Nice insight
Oh well
Interesting, but What proportion of the Pool Rewards is distributed by VoteBots would have been more interesting. I think it would be more than 25%.
Indeed, my first thought was "To how many $ do these correspond"?
However great thought, as always @penguinpablo, thank you!
very interesting, I thought it would be much larger amount
I thought it would be worse. Less than 2% of all votes isn't so bad. We need to keep an eye on that number, though.
Oi, that double axis graph is a little confusing at a glance.
Fascinating data. Thanks for compiling and sharing.
I wonder if more people don't use vote bots due to lack of knowledge, or if it comes down to lack of funds to buy the votes. I know very often the vote-bot doesn't return a positive ROI depending on your timing.
I'd be curious if there is a better solution to the problem of worthwhile posts and post discovery instead of vote-bots.