Currently the splinterlands has avg. 400k active accounts are playing daily.
How many of them are actually card owner?
How many of them are bots or rental accounts?
This is something popping in my mind which I really want to figure it out.
In order to figure out this question, here I will make some assumptions and definitions:
- Card owner account: The account holds most of his cards, and minority of the cards could be rented for some reason.
- Rental account or Bots: from the end user point of view, I have no way to really differentiate which account is rental and which is bots. Although many bloggers tried to differentiate the bots from human via their names. but for my personal point of view, differentiate from account name is not sufficient as people may create a series of accounts and put a run number on each account for sake of easy management.
- Assume the distribution of the bots, rental and human accounts play along different time period follows the uniform distribution.
- Assume the bots, rental and human accounts ratio along all league are same. (If I can find further data on different league, then we can refine this assumption.)
Therefore with the method that I'm thinking about, it will be only able to differentiate the card owner account from bots and rental accounts.
Here's the differences between Normal distribution vs. the Uniform distribution.
Gaussian distribution: Along the time, Bots, Rental, Card owner will mostly play at certain concentrated time period. and their period could be different from each other.
Uniform distribution: Bots, Rental, Card owner accounts distributed along different time period equally and among these 3 , the ratio of each of among samples remain same proportion.
Ok. Once we have the definition of different account and sample distribution type assumed.
Then how to derive the % between card owner account vs. bots/rental?
If the distribution assumption is true, we can get the ratio between card owner and bots/rental from the battle log that we have.
Here's the way that I'm going to do at least to estimate the ratio in Gold II league:
- Get the total battle log for my account ==> 20
- Check each opponent's card holding status.
- If the cards he holds roughly match to the CP he has ==> Card owner account
- If the account only hold rewards cards and the CP is significantly lower than the battling CP ==> Rental/Bots account
- After know the QTY of the card owner account vs. Bots/Rental, then we can get the ratio.
- with the PeakMonsters' dashboard plot, multiply the active account with the ratio.
==> Then I know roughly the daily number of account for Card owner vs. the Bots/Rental.
Understood, this is not a very accurate method to calculate the account number. But, this result does give some inside that how many accounts are actually putting money into the eco system and how many does not. And hopefully the new changes that will happen in early May 2022 will cause the changes on this ratio later.
Ok. Let's see how is the battle log looks like in my account
The CP that those opponents have vs. the actual est. CP they own are as below:
- NA-SPL-89: Battle CP:101,245, Actual CP: 2149 => Bots/Rental
- CALEBJ: Battle CP:512,456, Actual CP: 559,761 => Card owner
- RJMSO5: Battle CP:170,620, Actual CP: 0 => Bots/Rental
- PNHIRO: Battle CP: 24,754, Actual CP: 11,383 => Bots/Rental
- ALDREN3: Battle CP: 208,145, Actual CP: 230 => Bots/Rental
- THEGMACHINE: Battle CP: 347,530, Actual CP: 36,198 => Bots/Rental
- GIOTRIX: Battle CP: 277,197, Actual CP: 308,153 => Card owner
- HOCALL: Battle CP: 452,630, Actual CP: 777,949 => Card owner
- DC3NTR4LZD: Battle CP: 203,910, Actual CP: 236,847 => Card owner
- WYATTEARP: Battle CP: 2,195,519, Actual CP: 2,315,600 => Card owner
- JAMOOL: Battle CP: 155,400, Actual CP: 157,082 => Card owner
- DRAGONGHOST: Battle CP: 239,680, Actual CP: 0 => Bots/Rental
- Raijinxd091396: Battle CP: 0, Actual CP: 0 => Bots/Rental
- DORYVE: Battle CP: 173,352, Actual CP: 9,123 => Bots/Rental
- ANGELFEST: Battle CP: 599,645, Actual CP: 79,867 => Bots/Rental
- RJMS04: Battle CP: 0, Actual CP: 0 => Bots/Rental
- XANDER09: Battle CP: 0, Actual CP: 0 => Bots/Rental
- JJLEEPCMOTOVLOG: Battle CP: 0, Actual CP: 0 => Bots/Rental
- INSTAMENTAL: Battle CP: 374,830, Actual CP: 2,371,225 => Card Owner
- SKYMIN: Battle CP: 911,004, Actual CP: 2,106,838 => Card Owner
Ok. Now we got all the data between Card owner account and Bots/Rental accounts:
Card Owner: 12/20 (60%)
Bots/Rental: 8/20 (40%)
Avg. daily Active accounts: 401,750
So in these active accounts: 241,050 accounts are from Bots/Rental
and the remaining 160,700 accounts are the one who really put the money into the Eco system.
Caveats: This calculation is only from one day data, could only be used for reference. Tt may be far away from accurate.
Nevertheless, Since my account is at mid league - Gold II, so when you use this method to calculate, the result may be different when the league is different.
However, it does show a sad fact - Splinterlands has fair amount of accounts are running in minimum investment. (put money into the eco system.) In certain level, it's also kinds of unfairness for the people who really put their money into it.
To this point, different people have different thought and opinions.
Since there will be some new changes coming out to address the fundamental of those issues. Let's see how it will work.
Great insight, but I must tell you that "NA-SPL-##" accounts are scholarships account played by humans from NFTy Arcade Guild. Similar to "PD##" owned by PlayDough Guild, and "BLTZR-WIZARD-##" owned by Balthazar Guild.
So there needs to be another special category like "Scholarship Account" in your future calculation.
But nonetheless, great article. Thanks for sharing.
!PIZZA
Congratulations @splinterbin! You have completed the following achievement on the Hive blockchain and have been rewarded with new badge(s):
Your next target is to reach 50 upvotes.
You can view your badges on your board and compare yourself to others in the Ranking
If you no longer want to receive notifications, reply to this comment with the word
STOPTo support your work, I also upvoted your post!
Check out the last post from @hivebuzz:
Support the HiveBuzz project. Vote for our proposal!
PIZZA Holders sent $PIZZA tips in this post's comments:
@tedus(1/5) tipped @splinterbin (x1)
You can now send $PIZZA tips in Discord via tip.cc!
You math funny.
I applaud the effort here. But I don't think you can assume an equal distribution amongst the leagues. It's just easier to run bots in bronze and silver. In the higher leagues you see more of the customized bots. In bronze you see the generic bots people rent from the big suppliers. They can just play the ghost cards and rent any cheap card for CP.
There are also just a much larger sheer number of accounts playing bronze and silver.
Also, I now rent out $75k worth of cards. I can tell you that they rent in chunks all at once to numbered players and I'd say that over 90% of the cards I rent out are never played. Bots are probably over 300k of the active players. Just bots renting for CP
Probably the method as mentioned is very much simplified.
however, one thing makes it a bit trustable - I'm in the Gold II which is at the median position among all leagues.
I gave it another thought as what you mentioned, the distribution. I think the distribution for Bots, rental account and card owner account pretty much different. I would say something in below picture. it could also different from it depends on the profitability. since i did not see the bots/rental account at Gold II compared to Bronze, so i believe below distribution is still valid in a certain way.
Hello, you have the percentages flip flopped. CO is 8 and B\R is 12.
I like this "simple" method give some great insight. Roughly 50% bots seems reasonable. Even a bit low.
I think bots are online battling 24/7 were players are only playing a few battles. Mightbe the amound of bots is not as high as I thought.