Banlist Discussion!

[luapmox]

I prefer to not bad any cards unless there is more data made available. Having a greater variety of cards available to pull from makes for a richer game experience. I have lost with all of the cards mentioned in this thread and beaten them as well. Obviously my personal experience is going to be a very small sample which is why I need more data to be convinced banning is necessary. I agree with the argument that a few others have raised that there is always going to be a best card; if you ban it then the second best becomes the best. Luck is always going to be a part of this game and it is a huge part of what makes it fun. Sometimes you get the best card and sometimes your opponent gets the best card. In my opinion these cards do not end the game by themselves.

[Ranior]

I just want to clarify a few things from RicardoRix and luapmox's comments.

Nobody is arguing any of these cards are so strong that they win all the time. Of course all these cards lose sometimes. The criteria for a card that is too strong is not "it wins every game", that's not a sane or valid standard for what it would take to ban a card for being too strong.

Also, of course there will always be a card that is strongest, and then next on, and so on. That is not a reasonable argument either for if we should or should not be banning things for strength. It's a slippery slope fallacy. Banning a strong card does not mean that we cannot stop at some point, it's not like we start banning things for strength and then just keep going down the list forever. The question of where it stops is "somewhere". The whole point of this thread is to discuss where that line is, which I do want to hear arguments for.

But arguments along the line of "this card isn't an instant win" or "if we ban this strong card where does it stop?" are not logically sound arguments.


The question really comes down to if some cards win an aberrant amount. If we have a card that wins 90% of games and is always a first pick, I think we'd all hopefully agree that that should be banned. The game should not be determined by who gets lucky enough in the draft to win. Yes, variance is present in this game, sometimes you'll be in a worse draft spot or better, but that should only moderately affect your baseline percentage to win. Figuring out the exact line we're all comfortable with is going to be tricky, and this is where it would be great to have real stats on this, but at the moment it isn't clear we're getting updated statistics soon, and so we have to try and figure out what the line should be and what might currently be over it.

I don't know what the exact numbers should be, but my experience, and discussions with some others, has led me to think that all of Begging Student, Assistant Tiller, and Big Country in this present set lead to an unacceptable rate of 1sts and 2nds. These cards when played well rarely take 3rd or 4th, and I think that's a big problem. They are aberrations compared to even other strong cards--things like Carpenter's Parlor, Grocer, Wooden Hut Extender, Clay Supports, Bookcase/Bookshelf, Lantern House, etc I think do not have anywhere in the same realm of probability of helping someone achieve 1sts and 2nds as certainly as the three cards I am concerned with. (In particular Begging Student seems very aberrant in its winning percentage at the moment).

I don't know if it is possible for us to try and crowdsource some statistics ourselves, because I agree hard data would be the best for making these choices now that we have thousands of arena games to specifically look at for this set. However I personally lean on the side of banning things now that seem problematic--we can always remove stuff later if we do get data.

[RicardoRix]

I don't know what the exact numbers should be, but my experience, and discussions with some others, has led me to think that all of Begging Student, Assistant Tiller, and Big Country in this present set lead to an unacceptable rate of 1sts and 2nds. These cards when played well rarely take 3rd or 4th, and I think that's a big problem.

Just to make a statement about Big Country. You even mark this card with 'when played well'.
This would seem like a high skill card. Good players know how to win with it, lesser players don't and therefore won't draft it, and so the good players are far more likely to get it in the draft. I say 'far more' in comparison to other equally good cards but take less skill to use.
So there is a substantial swing here, that the card ends up in the good/winning players hand just alone for this fact. Even if it gets played or not!

Just for anecdotal evidence, I am not able to play Big Country, it just doesn't fit with my thinking, I either never play it, or make stupid plays to get 2VP out of it. In the games I have seen it being played and the player 'went for it', the player has came last. So you can take that from the ~200ELOish game value.

My evidence would support the fact you should keep it in, so good players can win with it and show us how good they are and teach us all something.
On the other hand if it's too easy to win with once the skill level increases, then I guess it's a different decision when you have all high skilled players at the table.
Shades of grey.

[bananasplay]

Just to make a statement about Big Country. You even mark this card with 'when played well'.
This would seem like a high skill card. Good players know how to win with it, lesser players don't and therefore won't draft it, and so the good players are far more likely to get it in the draft.

Just want to state that I appreciate the insight of this comment (which I've coloured to be clear about which comment I'm referring to).

My evidence would support the fact you should keep it in, so good players can win with it and show us how good they are and teach us all something.

I also appreciate the wisdom demonstrated here. If I've undervalued a card and my opponent beats me over the head with it, then I deserve to lose, but if I'm humble enough, I can also learn something.

On the other hand if it's too easy to win with once the skill level increases, then I guess it's a different decision when you have all high skilled players at the table.

Possibly. It may well be that the strength of individual cards may depend on the relative strength of the players at the table.

[Ranior]

Just to make a statement about Big Country. You even mark this card with 'when played well'.
This would seem like a high skill card. Good players know how to win with it, lesser players don't and therefore won't draft it, and so the good players are far more likely to get it in the draft.

Just want to state that I appreciate the insight of this comment (which I've coloured to be clear about which comment I'm referring to).

My evidence would support the fact you should keep it in, so good players can win with it and show us how good they are and teach us all something.

I also appreciate the wisdom demonstrated here. If I've undervalued a card and my opponent beats me over the head with it, then I deserve to lose, but if I'm humble enough, I can also learn something.

On the other hand if it's too easy to win with once the skill level increases, then I guess it's a different decision when you have all high skilled players at the table.

Possibly. It may well be that the strength of individual cards may depend on the relative strength of the players at the table.

So, to be clear, we did get some statistics from last season. The game data we got was pretty good. (Sadly we cannot at the moment get updated stats, there is some discussions with BGA/Lookout/others on who can get what data. Hopefully they will get resolved and we can get updated numbers). https://boardgamearena.com/forum/viewto ... 46&t=23574

However, the stats are there for last arena season for example. I guess last season is a bit of a misnomer, this is just a slice of like 2-3 weeks of game data, but still has quite a few samples. Among all skill levels, Big Country's average drafted position was 1.54. The highest ADP you can get is 1.00, that would represent it literally is always a 1st pick. 1.54 represents something like 50% of the time it gets taken 1st and 50% it gets taken 2nd in the draft.

One all players were at least 200 elo, that ADP goes even further to 1.33. These numbers do not in any way represent a card that is getting slept on in drafts and winds up in stronger players hands much more often. Maybe slightly more often as it does occasionally get passed, but rarely. Players pick it very highly, basically picking it before almost any other minor improvement option. (Carpenter's Parlor is the only other card that comes particularly close). Yet despite getting picked that highly, it still produces a very high chance to win (and those power calculations don't take in to account the odds of getting 2nd either). This is exactly why I am worried about cards like Big Country--I still find it gets drafted very highly and gets an aberrant number of 1sts and 2nds. Cards like that are big issues because you can't argue that people should just be drafting it higher, they already are drafting it very highly and yet they still are winning a lot of games by doing so.

(An example of the sort of card that does fit your argument: I'd say Milk Jug. For some reason Milk Jug tends not to be drafted particularly highly, yet it has a pretty good aid to winning [less than Big Country, etc, but still it's a strong card]. Yet its ADP is something like 3.59 in the all 200 elo set. That's a card that should start getting drafted higher probably, and then it's power will drop a bit, etc.)

Unfortunately we really don't have strong data on how cards fare in the CD environment, the slice of data we have is from alpha testing days when there were still bugs and a lot of players were still very much learning the card set. I'd be fascinated to know what the arena data shows from the 1000 games that have already been played this season to see what the average draft spot and rate of 1sts and 2nds are for the cards, but I'm not sure when we'll get hard numbers. My experience however does not match with yours--players draft Big Country with high priority still, and many put up very strong games. It is not a card you tend to see floating around for like a 3rd of 4th pick too often, especially not at higher skill tables.

[MisterG]

Me, I'd still like to know how Big Country is overpowered in 2p games, whereas Begging Student absolutely is (in my experience).

[EconSean]

It's been a couple of weeks of playing arena, so I'd like to update this thread with my thoughts on cards:

I now think the following cards should be banned:

Big Country: My analysis from the earlier post still holds. When played on two rooms it wins too high of a percent of the time according to the data we have, and it simply warps the game too much around it. We have enough data and anecdotal evidence to confirm that it is simply too powerful/format warping, and should be banned. I should note that two room strategies can be powerful without big country (with scholar/plow maker or even with collector into a late third room), but they will show up far less often.

Begging Student: The preliminary data we have suggests this card is powerful, but in my experience in Arena this card simply provides way too much value for so little cost. It's a very fun card to play with, but getting up to 6 actions at the cost of a beggar is just ridiculous. You can play this card with very strange occupations and still score very highly due to its raw efficiency. Add in that this card combos with occ cards, and that it enables you to play minors which normally have a high prerequisite, it's too much value to be printed on a single card. As a thought exercise, I think if this card gave 2 beggars it would still be one of the top occs. At 3 it would still be playable.

The following cards should be watchlisted:

Assistant Tiller: This card is powerful, and is ridiculous when combined with other day labor cards. However I do very rarely see it combined with day labor cards in the current set due to both card dilution and people picking day labor cards highly, and some combinations are not the worst to deal with (excavator, water worker, and cottager alone are beatable without a critical mass of day labor cards). Competition for day labor also hurts this card quite a bit, which can happen a fair bit. The floor on this card is high, and it is unbeatable with enough day labor cards, but I think that if we got updated stats the stats on assistant tiller would not show that it's overpowered in the current format. It is still incredibly powerful, and if the community decides that it's too powerful I don't mind that.

Potter Ceramics: This card does give you basically infinite food. There is a pretty real cost in that getting clay setup and an oven setup can be difficult in some games, but as long as you can do it there are many terrible cards that give bake actions that you will see because outside of this card baking using an occupation/minor improvement is seldom amazing. I do think the cost to setting this card up is fairly real in the early game, and as a result people rushing to set this up often find themselves behind in the growth queue. It also adds a lot of food to the game, making others' actions more efficient. However there are some games where this card is dumb (especially if you can grow at the same time or not have to fight for an oven), and so I'd be fine with this being banned until more cards are added to dilute the pool of mopey bake cards.

[Almicantarat]

I really do not understand what is the reason why we should ban cards.

I mean, this is an already REVISED EDITION, in which many adjustments to the game have been implemented.

It is really dangerous to ban cards based on gaming data which are biased. The bias comes from the amount of users that actually play the game here on BGA. Most of the games are played by a pool of 100 users maximum, which have their own styles, and since are playing many games per day, are playing not for pure fun but for winning.

And when you play for winning and you have so much experience, it is quite obvious that here people want to ban powerful cards in order to prevent less experienced users to win. I find this kind of behaviour quite deprecable.

[EconSean]

I really do not understand what is the reason why we should ban cards.

I mean, this is an already REVISITED EDITION, in which many adjustments to the game have been implemented.

It is really dangerous to ban cards based on gaming data which are biased. The bias comes from the amount of users that actually play the game here on BGA. Most of the games are played by a pool of 100 users maximum, which have their own styles, and since are playing many games per day, are playing not for pure fun but for winning.

And when you play for winning and you have so much experience, it is quite obvious that here people want to ban powerful cards in order to prevent less experienced users to win. I find this kind of behaviour quite deprecable.

Magic: the Gathering, which is a card game made by Wizards of the Coast, has frequently had to ban cards. Oftentimes it is because they print something that is too powerful in competitive play. Players in casual settings can still play with those cards if they choose to, but Wizards chooses to maintain a banlist for the health of the competitive format. Hasbro, which Wizards of the Coast is a subsidiary of, brings in 5.1 billion dollars in revenue, of which magic is somewhere in the ballpark of 5-7% of their sales. Needless to say, they dedicate a lot of funding to designing cards, rules, and formats in a way that is cohesive, and they frequently make mistakes. This is because they are developed by a team of, at most, hundreds of people, while the game is played globally by 20 million people.

The revised edition of Agricola, on the other hand, was developed by Lookout Games, which is an incredibly small company. The only estimate I could find is that the company has two employees, according to craft.co (unsure how reliable that is). Even though they relied on some community feedback from playagricola.com in order to make their decision, many of the key decisions were made by an incredibly small group of people, when now, through the power of the implementation here, the game can be enjoyed by thousands of players. These thousands of players, although they may not have been able to creatively develop Agricola, deeply understand the gameplay at a high competitive level, and as a whole I believe that we can sense when cards lead to unfun or unbalanced gameplay. Once again, bans are JUST a thing for competitive play. At a casual level, players can play with whatever cards they feel like they want to.

Even with balance, there exist many opportunities in Agricola for worse players to beat better ones. The purpose of a banned list is to provide balanced and fun gameplay in competitive play. If you are not a competitive player and therefore do not want to play with a banned list, I have great news, which is that in your games you can choose to not play with a banned list. Arena is a mode made for competitive players, and as such a banned list is appropriate to provide balanced and fair gameplay in which skill can be appropriately expressed.

[Almicantarat]

I still want to play competitive in Arena without any banlist.

I hope that the revised edition has been blessed bye Uwe himself before publishing (and as far as I know this is the case).

There is a huge difference between the creator and the gamers. The creator knows exactly the intrinsic value for each card/action. The gamers just know how their intrinsic value become more or less juicy in an opportunistic way, i.e. tactically.

Banning cards from the gaming community implies a forced coordination in the tactical way in which we should play the game, which restricts a lot the variety of the game itself. And variety is the real fun in agricola.