STEEM's Bidbot Business Income Analysis 2018 YTD

in #utopian-io5 years ago

Repository

https://github.com/steemit/steem  

 The bidbot business on Steemit is subject to much controversy, with both sides having a fair and reasonable arguments for their development and use on Steem.  Like it or not they are now part of the Steem ecosystem. 

3 months ago I produced a report on the bid bot business which you can read here

https://steemit.com/utopian-io/@paulag/steemits-bidbot-business-income-analysis-2018-half-1

However that analysis focused on SBD and a lot has changed on STEEM since.  For a while SBD was not being printed and the bots had been willing to accept STEEM for some time before that.

With that in mind, I have changed the reporting from SBD to Vests.  I have selected VESTS as all curation rewards are paid in VESTS (SP) and so is a portion of the Author Rewards.  It was therefore easier to convert everything to VESTS.

For conversions I have used the average SBD per STEEM rate for 2018 to date which is 0.99 and I have taken the STEEM per VESTS to be 0.000495.  As I have taken fixed values for the conversions, this report is an approximation.

The Time period on the report is Jan 1 2018 to Oct 15 2018.  As October is only a partial month, it may not reflect the true position of the month.

Aim of Analysis

The aim of this analysis is to establish 

1.  Descriptive statistical information for an overview on bids sent to bidbots in vests

2. Establish what % of Curation rewards are paid to bidbots

3. What % of Vests paid out in author rewards were sent to bidbots

4. Establish what % of Total Vest claimed goes to bidbots

5. Visualize year to date trends 


 Descriptive Statistics on bids sent

Descriptive statistics are useful for summarizing large sets of data.  The aim of this was to get an overview of the bids sent 

 It is interesting to see that median vests sent to bidbots has reduced from Jan to Aug and then increased again in Sept and now at an all time high in October so far.

 The standard deviation tends to vary month to month. A low standard deviation indicates that the data points tend to be close to the mean, while a high standard deviation indicates that the data points are spread out over a wider range of values. 

What % of Curation Rewards were paid to BidBots?

 The business model of the bidbots allows them earn curation rewards on posts they voted on in addition to the bids made.  Year to date 26.21% of Curation Rewards claimed are paid to bidbots.  The purple bar chart shows the total vests claimed from curation rewards this year, by month. We can see that Aug had the highest total vests curation reward this year. However while the total curation vests earned increased  from May to Aug, bidbot owners seen almost no movement.  

While looking at these charts, please remember only 15 days of October have been taken and these values could smooth out.

 The pie chart below shows this by bidbot 

What % of Author Rewards were sent to BidBots as payment?

Bidbots take payment in both SBD and STEEM.  Users have the option of using earning SBD and STEEM or purchasing it from an external exchange.  Its hard to establish exactly where the fund come from so for this part of the analysis I am taking the assumption that bot users are investing in bidbots with funds earning on authors rewards.

So what % of Author Rewards earning on STEEM this year went to bidbots?

we can see from the chart, the % of authors rewards that were ultimate used to pay for bidbots increased from Jan to a high in June 24.5% after this the % declined to 15.5% in September.  Although October seems to have risen, as only part of the month was taken this could still change.

On average for the year to date 19.01% of all Authors rewards claimed were sent to bidbots.

What % of Total Vests paid out by the rewards pool end up in the hand of Bidbots?

By combining the curator rewards earned by bots and the vests they received as bids we can calculate their revenue.  This can then be compared to the total VESTS claimed by both authors and curators.

We can see that year to date 19.5% of all vests paid from the rewards pool ended up in the hands of bidbot owners.  This is a combination of curation rewards they earn and bids sent to them.

The % increased from Jan to June and then decreased again to September.

Year to date Trends

The chart above shows the year to date trends for bidbot usage.  We can see the daily posts sent for promotion have reduced since May by 75% and we can also see clearly the impact of the HF20 roll out.

Conclusion

 The use of bidbots has been reducing and have fallen by 75% since its peak.  Comparing this to STEEM in general there has been an approx drop of 63% in the number of posts since the peak in bidbot use.  

The median vests sent per bid has also reduced over the year, however without full October figures its difficult to see if there is a break in the trend.

19.1% of all author rewards claimed ultimately ended up in the hands of bot owners

26.2% of curation rewards were paid to bidbots

A an approximate of 19.5% of all rewards claimed ended up back in the hands of bidbot owners.

Discussion

Please do comment below.

What do you think of the bidbot business in general?  Do you think it is good or bad for STEEM?  

The Data and Queries

First I used the Steembottracker API to get a list of bidbots.  This list was then used in the M queries belowThe data was collected from SteemSQL using PowerBI.  The following M codes were used.M code for Sent to bid bots

let    Source = Sql.Database("vip.steemsql.com", "DBSteem", [Query="Select #(lf)*#(lf)From TxTransfers (NOLOCK)#(lf)Where timestamp >=CONVERT(DATE,'2018-01-01')#(lf)and [to] in ('tainika' ,'chronoboost' ,'getkarma' ,'weupvote' ,'alliedforces' ,'a-bot' ,'boinger' ,'jerrybanfield' ,'upyourpost' ,'pwrup' ,'brotherhood' ,'haveaheart' ,'alfanso' ,'whalepromobot' ,'joeparys' ,'peoplesbot' ,'votepower' ,'t50' ,'moneymatchgaming' ,'stef' ,'sureshot' ,'ptbot' ,'th3voter' ,'edensgarden' ,'oceanwhale' ,'botox' ,'whalecreator' ,'profitbot' ,'ecotrain' ,'automation' ,'siditech' ,'cabbage-dealer' ,'profitvote' ,'steemerap' ,'ubot' ,'dlivepromoter' ,'emperorofnaps' ,'proffit' ,'lrd' ,'bodzila' ,'peace-bot' ,'brandonfrye' ,'authors.league' ,'flymehigh' ,'noicebot' ,'redwhale' ,'lost-ninja' ,'dolphinbot' ,'rocky1' ,'estabond' ,'minnowvotes' ,'thebot' ,'booster' ,'slimwhale' ,'megabot' ,'singing.beauty' ,'estream.studios' ,'dailyupvotes' ,'ebargains' ,'promobot' ,'honestbot' ,'foxyd' ,'sunrawhale' ,'mitsuko' ,'onlyprofitbot' ,'spydo' ,'isotonic' ,'brupvoter' ,'postdoctor' ,'luckyvotes' ,'therising' ,'inciter' ,'redlambo' ,'shares' ,'nado.bot' ,'bid4joy' ,'voterunner' ,'steembloggers' ,'upmewhale' ,'mercurybot' ,'smartsteem' ,'postpromoter' ,'upme' ,'msp-bidbot' ,'aksdwi' ,'pushup' ,'appreciator' ,'sneaky-ninja' ,'lovejuice' ,'minnowhelper' ,'boomerang' ,'buildawhale')"]),    #"Changed Type" = Table.TransformColumnTypes(Source,{{"timestamp", type date}}),    #"Filtered Rows" = Table.SelectRows(#"Changed Type", each Text.Contains([memo], "https"))in    #"Filtered Rows"

M Code for Claimed rewards

let    Source = Sql.Database("vip.steemsql.com", "DBSteem", [Query="Select * From TxClaimRewardBalances (NOLOCK)#(lf)Where timestamp >=CONVERT(DATE,'2018-01-01') #(lf)"]),    #"Changed Type" = Table.TransformColumnTypes(Source,{{"timestamp", type date}})in    #"Changed Type"

M Code for Bots curator rewards

let    Source = Sql.Database("vip.steemsql.com", "DBSteem", [Query="SELECT#(lf)reward,#(lf)timestamp,#(lf)curator#(lf)FROM VOCurationRewards (NOLOCK)#(lf)Where timestamp >= CONVERT(DATE,'2018-01-01')#(lf)and [curator] in ('tainika' ,'chronoboost' ,'getkarma' ,'weupvote' ,'alliedforces' ,'a-bot' ,'boinger' ,'jerrybanfield' ,'upyourpost' ,'pwrup' ,'brotherhood' ,'haveaheart' ,'alfanso' ,'whalepromobot' ,'joeparys' ,'peoplesbot' ,'votepower' ,'t50' ,'moneymatchgaming' ,'stef' ,'sureshot' ,'ptbot' ,'th3voter' ,'edensgarden' ,'oceanwhale' ,'botox' ,'whalecreator' ,'profitbot' ,'ecotrain' ,'automation' ,'siditech' ,'cabbage-dealer' ,'profitvote' ,'steemerap' ,'ubot' ,'dlivepromoter' ,'emperorofnaps' ,'proffit' ,'lrd' ,'bodzila' ,'peace-bot' ,'brandonfrye' ,'authors.league' ,'flymehigh' ,'noicebot' ,'redwhale' ,'lost-ninja' ,'dolphinbot' ,'rocky1' ,'estabond' ,'minnowvotes' ,'thebot' ,'booster' ,'slimwhale' ,'megabot' ,'singing.beauty' ,'estream.studios' ,'dailyupvotes' ,'ebargains' ,'promobot' ,'honestbot' ,'foxyd' ,'sunrawhale' ,'mitsuko' ,'onlyprofitbot' ,'spydo' ,'isotonic' ,'brupvoter' ,'postdoctor' ,'luckyvotes' ,'therising' ,'inciter' ,'redlambo' ,'shares' ,'nado.bot' ,'bid4joy' ,'voterunner' ,'steembloggers' ,'upmewhale' ,'mercurybot' ,'smartsteem' ,'postpromoter' ,'upme' ,'msp-bidbot' ,'aksdwi' ,'pushup' ,'appreciator' ,'sneaky-ninja' ,'lovejuice' ,'minnowhelper' ,'boomerang' ,'buildawhale')"]),    #"Changed Type" = Table.TransformColumnTypes(Source,{{"timestamp", type date}}),    #"Replaced Value" = Table.ReplaceValue(#"Changed Type","VESTS","",Replacer.ReplaceText,{"reward"}),    #"Changed Type1" = Table.TransformColumnTypes(#"Replaced Value",{{"reward", type number}}),    #"Renamed Columns" = Table.RenameColumns(#"Changed Type1",{{"reward", "reward VESTS"}})in    #"Renamed Columns"

M Code for  Total Curation rewards

let    Source = Sql.Database("vip.steemsql.com", "DBSteem", [Query="SELECT#(lf)r eward,#(lf)timestamp,#(lf)curator#(lf)FROM VOCurationRewards (NOLOCK)#(lf)Where timestamp >= CONVERT(DATE,'2018-01-01')#(lf)"]),    #"Changed Type" = Table.TransformColumnTypes(Source,{{"timestamp", type date}}),    #"Replaced Value" = Table.ReplaceValue(#"Changed Type","VESTS","",Replacer.ReplaceText,{"reward"}),    #"Changed Type1" = Table.TransformColumnTypes(#"Replaced Value",{{"reward", type number}}),    #"Renamed Columns" = Table.RenameColumns(#"Changed Type1",{{"reward", "reward VESTS"}})in    #"Renamed Columns" 


Sort:  

Hi @paulag, great follow-up to the previous report! October seems to be again a strong month for the bots, but we'll have to wait for the data before drawing final conclusions. From what I've read elsewhere, vote buying profitability had better times than now. I haven't had a chance to look into it myself - did you have a look by chance? The changes to curation rewards are (or will be) interesting with HF20. While it previously was highly profitable to have a bot vote shortly after post creation (with close to 0 curation rewards for the bot), this advantage is now gone with HF20. Authors could maximize their rewards by having the bot votes shortly after the selfvote, but at least several minutes after creation - and this automatically means also better curation rewards for the bots.

Your contribution has been evaluated according to Utopian policies and guidelines, as well as a predefined set of questions pertaining to the category.

To view those questions and the relevant answers related to your post, click here.


Need help? Write a ticket on https://support.utopian.io/.
Chat with us on Discord.
[utopian-moderator]

I think for October I only took the first 15 days so its a partial month. I wanted to included it to see if HF20 had any major impact but other than that I would have excluded it.

I have not looked into the profitability of vote buying, but I have also heard that its under a bit of 'pressure'

What would be interesting to add would be a follow up on the post you did a few months back on what portion of the rewards pool bots and autovoting distribute, I wonder has that changed much now?

Thank you for your review, @crokkon!

So far this week you've reviewed 2 contributions. Keep up the good work!

I thought it is much worse than this.

silver lining then?

I don't think they are good for Steem in the long term and i don't think paying for votes on a social media site encourages the overall quality of posts. I am happy to see that their use is falling off.
I can understand that it is an attractive idea while the trending page is still the first page that people see when they enter Steemit being as it is an advertising method. but I believe that it creates a very bad image of what can be found here to newcomers. Seeing a post earning in the hundreds only to open it and find it is shilling a service is not going to help with adoption in the long run..

I understand this point very well. Trending is not a true reflection of what really happens.

I don’t find that number they are making as as nearly as shocking as the number bidbots of listed in your source. They sure can be used to get attention for whatever is the case.

Such as Netcoins as we have seen over past few days. I found it interesting the site stated a “$30k value” for first place. Then thinking just how much voting was going into many posts to promote it into trending and other places. They sure got a good deal to get word out about them. Which makes me wonder how much was spent to promote them compared to actual first place and if we even win that instead of a 25% off coupon.

there are a serious amount of bots, but some are way more popular than others.

Thats actually really interesting, I might take a look at that and see how much was given out in votes on these posts :-)

I personally do not use bidbots except for a couple of experiments I tried in the past. I do feel they have a place in the ecosystem for two reasons: (1) a better way to promote initiatives and posts that need attention and (2) a place where passive investors could feel they are getting an ROI for they stake. While both can be abused, I think that is why there are a number of high profile accounts out there to deter abused with blacklists or just by calling out actions of the users. However, looking at the data you provide, I think the balance has shifted somewhat towards a better spot were value can be created from all participants. I think only time will tell.

I am not a bot user myself, but never say never. I think if I had something I wanted massive visibility on I would consider it. there are a lot of pros and cons and its something I might discuss in a follow up post

I am in favour of a certain amount of automation with respect to curation (e.g. @steemauto).
I do struggle with the idea of delegating SP to some bot in order to maximize revenue.
That is like momentum trading in the stock market . It has been shown to be reliably profitable in the public markets. But it's riding the coat tails instead of discovery.

With respect SMT it could be fatal, since the emphasis will be on money (click bait, shocking, sensational), instead of curation (quality, research, relevancy).

That doesn't mean there should be no money in it. Just that the sole focus on money seems short sighted and might impede the long term success of the platform - which could probably be way more profitable in the medium term.

In short: do you want Huffington Post or quality content?

Love this reply, thank you for adding your opinion. I too am in favor of a certain amount of automation, but nothing beats human curation and engagement.

Bots are an excelent way to autofinance usser with poor social engagement (like me)

Ill put good effort and time on my post, and maybe nobody reads me because ill talk about politics, now im writing in english and its a little better.

But, be real, no body cares about politics... So bots help me to keep working on the original content creation (wich is the real value of steemit)

But to talk about politics its "banned" for some trash bots (or market votes) like @minnowbooster (if u guys read me, you suck) and they spread they Terms to other bots, so im kinda screw C=

And with my little SP now i cant comment not much since fork =/

Its like @Steemit doesnt love me =P

I just had a look at your account as the RC problem for newcomers is still a bit of an issue. But I found your not a new comer and had plenty of time to grow your account.

As for the politics of how these guys run their services, well its not really something I am interested in discussing to much, but if you dont mind me asking, does this 'ban' just effect the use of bots? because if so, see it as a good thing and grow your account organically. Find an audience that is interested in your content and promote it to them. Don't rely just on organic steem traffic, its hard to have a niche here that's not about crypto and be successful with traffic. Sometimes I think of steemit as wordpress, if I had a blog there, I would have to find my own readers. Might be worth thinking about?

ill drain my account a few monts ago, and when i start posting again after the fork screws me so bad! hehe! I grabbed some crypto I had on the exchange and passed them so I could write. I had 15SP (as if I was a new user) and I couldn't comment or post. I made a post and couldn't even answer the comments, it was horrible.

I have no problem talking about Bots. But I was banned, specifically for writing about politics (it appears in the terms and conditions of Minnowbooster) I think, as an opinion, that obviously the ideal is to find people who are interested in your topics. But, for example, when you speak Spanish as a native language the public you access is very small, and if we have curators who can make a significant vote (Economically speaking) is even smaller. Then talking about something that few are interested in (like Argentine news) is reduced even more. Should I change my niches? I think diversity is a plus point for continuing to do so even though organic growth is lower, because creating this diversity of topics makes Steemit more likely to appear in external searches and attract new users to register with Steemit.

And since I don't live on Steemit I don't have the time to interact as much as I would like. Especially because I like to expand on my comments. Apart from that, I think I touch on topics (within the political sphere) that nobody talks about, such as the importance of independent media. Corporate manipulation of information (since all newspapers, radios and TV channels are bankers and investment funds) among other analyses that I consider valuable. I also usually write a lot about philosophy, but English is hard enough for me and to talk about something as delicate as philosophy I don't think I'm ready. I put a lot of effort in what I write and also do my own design as you can see if you navigate in my profile, all the images are my authorship (in fact if you would like some do not hesitate to ask me).

I find interesting that their curation rewards percent raised sharply in October, compared to a steady decline during the previous months.

Did they adapt to HF20 curation changes earlier than manual voters? Or what's the explanation for that?

As October is only a partial month, I wouldn't be too concerned as there is a chance it will smooth out by month end.

Nice analysis!

Now I wonder if we could make bigger profits when using bid bots?
Cause when less Steemonians use the bid bots, the ones that are using them should get higher rewards. Unless the total upvote value from bid bots is decreasing.

Hi @paulag!

Your post was upvoted by @steem-ua, new Steem dApp, using UserAuthority for algorithmic post curation!
Your UA account score is currently 6.908 which ranks you at #95 across all Steem accounts.
Your rank has dropped 1 places in the last three days (old rank 94).

In our last Algorithmic Curation Round, consisting of 270 contributions, your post is ranked at #11.

Evaluation of your UA score:
  • You've built up a nice network.
  • The readers appreciate your great work!
  • Good user engagement!

Feel free to join our @steem-ua Discord server

Coin Marketplace

STEEM 0.36
TRX 0.12
JST 0.039
BTC 69965.85
ETH 3540.49
USDT 1.00
SBD 4.71