Jobs Support Register

Automated Balancing by Popularity

+
What if cards are being balanced based on their popularity (or lack thereof), instead of their strength? And what if this process is automated? Every week all popular cards (>60% usage) gain one provision and all unpopular cards (<40% usage) lose one provision and everything in between remains unchanged. What would happen? Do you think the meta will continue to shift, indefinitely? Or do you think that all cards eventually end up in the neutral zone (between 40% and 60% usage)?

To answer my own question, I think the meta will continue to shift. On the bottom end, "junk" cards are continuously being cycled when others become cheaper up to the point that everyone starts using them and they become too expensive again. Similarly, at the top, "OP" cards will become too expensive and they are going to be replaced by cheaper cards.

In other CCG, automated balancing would be impossible because cards do not have a generic/universal value (in quality, not rarity). The closest thing would be the mana cost. However, that is not a dial that can be turned because adding or removing one mana has too much of an impact on the card. Gwent, on the other hand, has the provision system, which would be perfect because provisions are a generic value, one that can be used across all cards equally.

PS. I am not actually suggesting Gwent should implement this (hence the thread is not in suggestions). No, it's more of a thought experiment.
 
Don't they already do this? Oh, this isn't the joke thread....

I don't think it would work out very well. The problem with card usage statistics is they omit context. Card A is getting used a ton. Nerf it. Card B is given the bad card label and nobody plays it. Buff it. Why? How? Why are these cards used frequently or not at all, respectively? How should they be strengthened or weakened? Context is everything.

The second obvious issue is an extension of the why question. Card A may see a lot of play because the value relative to the provision cost ensures it's always economical. Card B might be bad because it needs so many other cards included and drawn together to be worthwhile. In the first case a provision adjustment could potentially fix the issue. In the second case it's likely not going to do so. Likewise, +1/-1 provision may not be enough to change whether a card is good or bad.

The ideal way to balance the game would be to understand those two earlier mentioned questions. Why is a card too weak or strong? Either in general or within the meta once it settles. How could it best be changed to bring it in line? This requires a very good grasp on the card pool, meta and game. Going off card usage statistics itself is a recipe for failure. It would be like declaring coin flip fixed because the stats say it's "even".
 
Other CCGs have other ways of changing the meta, like only using a set amount of expansions for example. Gwent cannot really do this though as it has, in my opinion, not even enough cards to call it a base. Or maybe it would if half of them werent useless.
The issue with a balancing system like you suggest is sometimes, cards are not used for just their value. Remember in beta, when Shani and Slyzard got the -1 hammer? How did that affect their playrate? The answer is it did not at all, they had still playrates of over 90% because they a) provided too much utility and b) there were no cards to take their spot, even though they were weakens. Now, I am not playing right now, but I'd be willing to bet the Witcher trio has a similar thing going on?

Different question: im your system, if you would continue to ignore a card, would it eventually go down to 0 provision?
 
I don't think it would work out very well. The problem with card usage statistics is they omit context.
You are right about that. The correct way to balance cards is to analyze why they are being played too often or too little and tweak them individually. My suggestion (thought experiment) isn't a fix for that, but rather an automated process to create at least some offset while also keeping the meta from growing stale.

Likewise, +1/-1 provision may not be enough to change whether a card is good or bad.
The issue with a balancing system like you suggest is sometimes, cards are not used for just their value. Remember in beta, when Shani and Slyzard got the -1 hammer? How did that affect their playrate?
If one provision doesn't make a difference, then two provisions will, when they are added or removed in the next week's cycle. What I do see happening is that too expensive cards are being switched out, only for those to return with the next cycle when their costs are reduced again, resulting in an infinite loop between two meta states.

Different question: im your system, if you would continue to ignore a card, would it eventually go down to 0 provision?
Yes, but I don't think it will ever reach this phase because too many players are going to be using the card before that, meaning the provisions will go up again.
 
I find the idea interesting. But i would go for >70% and <30%. And if thi would be autoatic, i would preferre an alternating algorithm. Means 1. Buff after two weeks 2. Nerf, after 2 weaks Buff, after.... At this rate you can try some more but still have the sam time span for each card buffeed or nerfed. Otherwise it would be to much i think.