To whoever it may concern...
announced that the team is planning to remove the opponent name during ranked combat in an attempt to fight bots from exploiting auto surrendering.Last 27th of January @jabapmatt
My personal take on it
Opponent adaptation is an absolute fundamental of the meta-game of any card game.
I am totally against removing the name of my opponent, to me that is like trying to kill flies with a howitzer and it could have profound repercussions. Essentially, I believe there are other ways to penalise auto surrendering bots without incurring in dumbing down the game. Let's not forget that every card game has this aspect of adapting to opponent.
This game is advertised as a dapp or "decentralised app", which is like virtue signalling for transparency. Now, how transparent is blocking opponents names? How decentralised is having a central server temporarily obfuscating opponent names?
First you removed the battle data from HIVE blockchain and now this .. seriously guys you can do better than that. (I understood the technical reasons behind the decoupling of HIVE blockchain but I just don't see this).
Again, here are other ways to penalise win trading, examples:
Account A surrenders to Account B if this happens more than 3 times you ban both accounts (more than 3 times in a timespan of 1 month)
If one account is surrendered to more than the typical deviation it should get blocked (you can detect that statistically)
Just because the official UI says little about your opponent does not mean player info should not be exploited. If what you want is to create a more fair battleground, do it by enriching the experience instead of limiting it. (Create more advanced UIs for real players).
PD: free information will always be exploited better by smarter or better organised individuals/groups, you cannot change that. Now ask yourself this, you really want to combat free information?
Thanks for reading.
Cheers,
lightproject
This would be the more centralised approach. I think the win trading is less of a problem than the card ownership and battle history analysis of which bots are capable.
Within a fraction of a second, a bot can discern which team I'm least capable of countering.
Any time information confers a competitive advantage on bots, we should obscure it wherever possible.
I appreciate your passion for the game.
All of the card ownership and battle history is openly available information on the blockchain.
IMO a better solution is to make that information more available rather than dumbing it down. There is already an alpha free to use app that is doing this. https://risitasapp.splinter.monster
I agree with you that bots should not have any advantages over normal players, but I think enabling normal players with better tools/information is the best path to take.
So instead of your idea, "Any time information confers a competitive advantage on bots, we should obscure it wherever possible."
I propose "Any time information confers a competitive advantage on bots, we should make that info available to normal players wherever possible."
Information obfuscation is never the answer. Splinterlands is a blockchain game and advertises itself as a decentralised application.
BTW 99% of bots do not even bother to look at opponent tendencies.
There are ways to counter bots. One way is to create advanced user interfaces for real players so that they can have as much information and data as bots do.
I am actually working on an alternative UI with such stuff, I made a presentation post here.
This is classic gametheory. Instead of removing this element. Make a counter incentive. I.E. If you surrender you got a 10 minute timeout (this is very punishing for bots).
To say it simply. Surrendering should be more punishing than losing.
That would be easily exploited. You could just use a single card to induce an easy loss.
I mean, those bots surrender to make the game end quickly. Run an average calc on "average time a game takes typically. And then make that the timeout between games.
@ocdb may we know the reason for your downvote?
It was overrewarded due to the haejin/rancho combo.
Over-rewarded based on which criteria? I can personally find many posts in the trending "over-rewarded" but that is just my subjective opinion. Do you make use a formal protocol to evaluate what is considered a fair reward based on content quality?
Thanks for taking the time to reply.
Yeah we have certain criteria's but we don't think it's worth pointing them out in this post which is also why we didn't leave a comment to begin with as we thought it would be a bit of a given why the downvote was cast. If you look at rancho's voting activity you can tell how little regard he places onto what he votes, that said if there was a lot of effort and quality behind it and it would be a first time vote on an author or not having landed on the same account in a while (such as you) we would have ignored it but both lack of engagement after trending for a while and there not being that substaintial effort or quality or content in this post we figured a small downvote to remove part of rancho's strength was fitting.