What is the best time of the day / week to use upvote bots?
What Are Upvote Bots?
For better or for worse, upvote bots are a great way for us minnows to get our posts promoted and grow our follower base. For those of you who may not be familiar with what upvote bots are - they are Steem accounts (typically with a large amount of Steem Power) that you can send some SBD to and include a link to your post in the memo and they will then use a portion of their Steem Power to upvote your post.
Most upvote bots work on a bid system where every 2.4 hours (the amount of time it takes your voting power to recover from a 100% upvote) they split a 100% upvote between everyone who sent SBD based on the amount sent. So for example if a total of $100 SBD was sent in the last 2.4 hour period and you sent $10 SBD to promote your post, then your post will get a 10% upvote from the bot.
How to Maximize Your Returns
In order to get the maximum return from your investment, you want to send your bid in during a period with the least amount of other bids sent, which gives you a bigger slice of the vote pie.
I've been diving in to the Steem JavaScript API recently and thought it would be a useful exercise to analyze bids sent to the popular @booster voting bot over the past 30 days and see what times of day and days of the week the average bids were the least - meaning they present the best times (from an ROI perspective) for you to use them to promote your posts.
So using the API I pulled the activity of the @booster voting bot for the past 30 days, split it into the 2.4 hour slots between its votes, and totaled the bids sent during each period. Then I copied the results into Excel and whipped up the two charts below:
The first chart shows that Sunday is on average the best day to use the @booster voting bot since average bids are the least on that day of the week. The next best day is Saturday followed by Tuesday, Friday, Wednesday, Monday, and Thursday.
The second chart shows the average bids by hour of the day (GMT time). The best hours (lowest average bids) seem to be 2 am - 3 am and 11 am - 12 pm although in general the whole period from 2 am through 12 pm GMT seems to be pretty good with the odd exception of 7 am - 8 am which has an unusually high number of bids.
So overall it seems that the best times to use the @booster voting bot are roughly between 2 am - 12 pm GMT on the weekends. Hopefully this information is helpful to some of you, and I will continue to post any other insights I can think of as I delve into the Steem APIs even further - so stay tuned!
Great analysis. Is this still relatively accurate do you think? I imagine if and when demographic changes to the steem block chain occur, as mediated by the SP of the group, the times may change.
That suggests that the bulk of the users are posting during the inverse times, which indicates their available time to do so, outside of work or other commitments.
Quick query: you referenced GMT, just wanted to clarify if that is Greenwich local time without modification, ie: GMT+0, where NYC is GMT-5, PEK is GMT+8, SYD is GMT+11...
Thanks for sharing your work and insights.
Yes, one other thing I would like to look into is average overall post rewards by day and hour to see if there are times when more people are voting on posts, or potentially more whales at least.
Regarding your query you are correct about the time zone used. One thing to note is that NYC is actually GMT-4 right now due to daylight savings time.
I didn't want to complicate GMT question with DST :)
I too am interested in the possible correlation between time of post, number of view/votes and possible influence on final payout. Timezones impact when posts are made, and seen. As you know a post will very quickly become buried - if it doesn't find an audience of voters fast it is doomed, no matter how much work or value it may represent, whereas a lol-meme or nsfw image posted 1 hour later may receive votes an order of magnitude larger.
good information. thanks for sharing @yabapmatt
Glad you liked it - let me know if there's anything else you're interested in seeing and I'll try to include it in a future post!
Blessing @yabapmatt ...thanks for this interesting and useful information... Upvoted... Blessings
Nice insight. Thanks for sharing.
I wonder if this is a consistent trend with the other upvote bots.
It’s probably similar but I’m working on a post with more detailed information and including more bots so stay tuned!
Thanks for this, most useful analysis. I haven't yet started using any of the bid bots - they have somehow passed me by.
I notice you use them a lot on your posts. Are they always profitable or is there an element of luck involved? (presumably lessened with your intel above?)
The bid bots are definitely NOT always directly profitable. But even if you get an upvote for less than you paid it can get your post listed higher on the "hot" or even "trending" pages for the categories that your post is in. That can get your post seen by more people and get you more real upvotes which can make the bidding bots indirectly profitable.
That can be hard to judge and I'm working on a post with more in-depth information about how to use voting bots profitably so stay tuned!
Hi @yabapmatt I'll be really interested to read your results looking into this. -dt
Nice work Matt, I wonder how did you iterate the history of transfers? I found iterating on past blocks super slow using the Python client.
I just used the steem.api.getAccountHistory() method of the JavaScript API. Not sure how it works under the hood (though I would like to learn) but it just spits out all of the transactions for a specific account pretty quickly and then I just loop through them and grab all the ones of interest.
Very interesting insight @yabapmatt,
I was wondering about that myself. Thanks for sharing :)
Glad to help - I think upvote bots are a great way to help newer users get their content seen but when the amount of total bids in a period far exceeds the voting bot's 100% vote reward then the bot is just making extra profits at the expense of the users.
I think there's a happy medium between voting bot profits (which are important to keeping them available and running!) and ROI for users which is what I'm working to find for myself and others.
I very much agree with you.
I tried @booster and @bellyrub, but the payout was not enough to suffice the payment for it. Especially when we don't have enough SBD to waste.
I would like to try the times you suggested and see what are the results. Will let you know, if you are interested, not that my results will add much value to your stats.
Thanks a lot for the reply and for your work to help minnows :)
Yes, especially with @booster you can't expect a direct positive return since it's so popular but the idea is that the upvotes will get your posts seen by a larger number of users so you'll indirectly get a positive return overall from it.
Hi @yabapmatt
When you said:
What do you mean?
I dont see how my post being upvoted will make my post seen by large number of users!
Your post will show higher in the “hot” and “trending” pages the more votes/rewards it gets so more people will see it who aren’t already following you.
ok, I understand. Thank you for your reply.
However, I don't think it is that easy to reach "hot" or "trending", unless I am using large SBD amount to get large upvote. Otherwise, it would be minimal.
By the way, I tried to bid on @booster twice yesterday around 3am. But my bid was not accepted. My guess is that the amount I used was not enough (1 and 0.5 SBD)!! Although before I used similar amounts and it was ok!
Just thought to let you know :)
Nice analysis! I will try to get votes at these suggest times and see how much more value I can get.
This post has received a 5.71 % upvote from @lovejuice thanks to: @yabapmatt. They have officially sprayed their dank amps all over your post rewards. GOOD TIMES! Vote for Aggroed!