Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Morgrim7]

I would wait, unless a high proportion of people have bought Dark Ages...

referring to this, a good amount of people do own DA.

[Robz888]

I plan to brazenly rank the Dark Ages cards, despite having only played about 15 games with them.

Let's all theory-rank the Dark Ages cards! That'll be a lark.

In my experience, Mystic is a useless Card. I buy it--and win with it--less than Saboteur. Worst $5 card.

[Beyond Awesome]

I plan to brazenly rank the Dark Ages cards, despite having only played about 15 games with them.

Let's all theory-rank the Dark Ages cards! That'll be a lark.

In my experience, Mystic is a useless Card. I buy it--and win with it--less than Saboteur. Worst $5 card.

Mystic is okay. It is a non-terminal silver which at least puts it up there with Stash, but actually Mystic is more useful than Stash and with deck Manipulation like Wandering Minstrel or Cartographer, you can build an engine out of it. But, yah, you're right, I have found Mystic to be on the lower end of $5 cards.

[Piemaster]

I agree with doing it now with Dark Ages for most of the reasons already stated, namely:

1.  Otherwise the rankings will be too similar to the last ones.
2.  Because it doesn't matter if cards get over/underrated to start with, that's half the fun and
3.  Because Dark Ages is (likely) never coming to Iso, so it's not like we'll suddenly develop a greater understanding of it later on.

[timchen]

Is there any sort of ranking algorithm or something that would let someone rank just one set and still have their ranking considered?

In other words, suppose I only play IRL and only with the Base set, but someone shows me this site and the wiki, and I think the card rankings are really cool and I would like to participate in making the next list, but I only know the Base cards. Would it be possible for me to rank only the Base cards but still have my ordering influence the overall ordering?

I was just thinking this might encourage more participation; ranking cards is easier if you know all the cards that are being ranked, and not having played with all the sets might hold some people back.

Of course if there's no sensible algorithm for combining partial results this wouldn't really work; does anyone know of such an algorithm, or where to look for one? Anyone else think this is a good idea?

Interesting. It think this can be done. I haven't thought about this.

I've been thinking about a couple different ways to do this, the best idea I've had so far is to basically use the ELO system. The ranking list will be translated into a set of card vs card "games," where in each game the "winner" is whichever card the user ranked higher.

So using the last $6+ Cards list as an example, Adventurer would have 13 losses, Grand Market would have 11 wins and 2 losses, and so on, and each card's ranking would be adjusted with each "game", and at the end sorting by their new ELO ranking should give the same ordered list. Unfortunately this would take O(n^2) time, since each list of n cards makes (n choose 2) = n*(n-1)/2 "games" that need to be played between the cards.

This would mean lists submitted by users evaluating all the sets would have more weight (more cards in the list, so each card gets more "games"), but there is no obligation for the user to evaluate every set. (though of course for the fun of it we should strongly encourage ranking every set the user has played with, particularly so we can see the Dark Ages rating change over time)

This is way too complicated. I would say the easiest way is just to rescale the rankings (or if you play bridge, matchpoints in percentages that is.) Basically a player just ranks among the cards he wants to rank. The score a card gets is the (rank-1)/(n-1), where n is the total number of the cards he ranked.

This method is problematic when the player chooses a subset which is in average substantially weaker or stronger, or when he mixes a relatively strong/weak card along with a bunch of average cards. But I guess as a fun list it should probably be okay.

[Dsell]

I plan to brazenly rank the Dark Ages cards, despite having only played about 15 games with them.

Let's all theory-rank the Dark Ages cards! That'll be a lark.

In my experience, Mystic is a useless Card. I buy it--and win with it--less than Saboteur. Worst $5 card.

I bet you haven't played Prophecy, the recommended set for Intrigue and Dark Ages.

Armory, Ironmonger, Mystic, Rebuild, Vagrant, Baron, Conspirator, Great Hall, Nobles, Wishing Well

[Davio]

Let's do it with Dark Ages if only to look back in a year and see how wrong we were!

[kn1tt3r]

Dark Ages has been released for almost half a year now, so it would be quite pointless to do it without those cards. I understand that this forum is quite dependend on isotropic, but it's not the ultimate wisdom. Most people here have been able to play with those cards, so we should go ahead and rank them, even if we're not completely familiar with them.

[kn1tt3r]

Is there any sort of ranking algorithm or something that would let someone rank just one set and still have their ranking considered?

In other words, suppose I only play IRL and only with the Base set, but someone shows me this site and the wiki, and I think the card rankings are really cool and I would like to participate in making the next list, but I only know the Base cards. Would it be possible for me to rank only the Base cards but still have my ordering influence the overall ordering?

I was just thinking this might encourage more participation; ranking cards is easier if you know all the cards that are being ranked, and not having played with all the sets might hold some people back.

Of course if there's no sensible algorithm for combining partial results this wouldn't really work; does anyone know of such an algorithm, or where to look for one? Anyone else think this is a good idea?

Interesting. It think this can be done. I haven't thought about this.

Maybe you could just weigh their voting by the amount of cards they included. So if someone only ranks the sets without Dark Ages, roughly 82% of the total number of cards are considered, making also his vote only be worth that much. For Base-only it would be about 13%.

[Qvist]

Is there any sort of ranking algorithm or something that would let someone rank just one set and still have their ranking considered?

In other words, suppose I only play IRL and only with the Base set, but someone shows me this site and the wiki, and I think the card rankings are really cool and I would like to participate in making the next list, but I only know the Base cards. Would it be possible for me to rank only the Base cards but still have my ordering influence the overall ordering?

I was just thinking this might encourage more participation; ranking cards is easier if you know all the cards that are being ranked, and not having played with all the sets might hold some people back.

Of course if there's no sensible algorithm for combining partial results this wouldn't really work; does anyone know of such an algorithm, or where to look for one? Anyone else think this is a good idea?

Interesting. It think this can be done. I haven't thought about this.

Maybe you could just weigh their voting by the amount of cards they included. So if someone only ranks the sets without Dark Ages, roughly 82% of the total number of cards are considered, making also his vote only be worth that much. For Base-only it would be about 13%.

That's basically what I'm planning to do.

[Tables]

I plan to brazenly rank the Dark Ages cards, despite having only played about 15 games with them.

Let's all theory-rank the Dark Ages cards! That'll be a lark.

In my experience, Mystic is a useless Card. I buy it--and win with it--less than Saboteur. Worst $5 card.

In my experience, all of Dark Ages is terrible. I'm pretty sure I've won FAR more games with just about every other card in my deck on Isotropic than I have with any Dark Ages cards.

[brokoli]

In my experience, all of Dark Ages is terrible. I'm pretty sure I've won FAR more games with just about every other card in my deck on Isotropic than I have with any Dark Ages cards.

Is it ironic ?
In my experience, Dark Ages cards are all pretty good.

[dondon151]

In my experience, all of Dark Ages is terrible. I'm pretty sure I've won FAR more games with just about every other card in my deck on Isotropic than I have with any Dark Ages cards.

In contrast, if I've won 1/1 games with Dark Ages cards, then they are teh best!

[() | (_) ^/]

In my experience, all of Dark Ages is terrible. I'm pretty sure I've won FAR more games with just about every other card in my deck on Isotropic than I have with any Dark Ages cards.

Is it ironic ?
In my experience, Dark Ages cards are all pretty good.

Hint: Dark Ages isn't on Isotropic.  Reread his last sentence with this in mind.

[brokoli]

In my experience, all of Dark Ages is terrible. I'm pretty sure I've won FAR more games with just about every other card in my deck on Isotropic than I have with any Dark Ages cards.

Is it ironic ?
In my experience, Dark Ages cards are all pretty good.

Hint: Dark Ages isn't on Isotropic.  Reread his last sentence with this in mind.

Lol, ok !

[Davio]

The best part about new expansions for me is the oscillation that I have with every card, from "It's terrible!" to "It's fantastic!" to "it's so-so" and ending up with some understanding of the card. And then again that understanding is only limited to the kingdom-types I've played with it.

[Turambar]

I've never submitted to these mainly because I don't know the criteria to use. Many of the discussions after the lists are announced goes:

"Card X is better than Y because it is good in every game, while Y is only good sometimes."
"Card Y is better than X because when it is good it is game-changing, while  X is just always decent."

These persons agree on the card, but disagree on criteria. I will not be submitting this year either, unless there is some strict criteria for ordering the cards.

[DStu]

These persons agree on the card, but disagree on criteria. I will not be submitting this year either, unless there is some strict criteria for ordering the cards.

I don't think there is any reasonable simple criteria to use.  You have these at least two dimensions, and if you just order according to one of these dimension you don't get anything that has any meaning.  A card that is in 99.9% (note: arbitrary treshold) of the games a bit more powerful than Silver (note: ...)  is not the most powerful $3, and a card that gives you over 9k VPs (note: ...)  in 0.001% (note: ...) of the setups.

I think the discussion afterwards is always very fruitfull.  It's not for reaching an agreement on what is stronger, but to point out the powers/weaknesses of a card.

[jonts26]

I typically assign a power level to each card over a certain percentage of games and do some sort of weighted average. Well I don't actually pick numbers and do the math, but I guess I could. It's just my general criteria.

So for instance if 0 is detrimental to buy, 2.5 is indifferent, and 5 is must buy - Coppersmith might be like 0 60% of boards, 1 on 15%, 2 on 10% and 3 on 7% and 4 on 5% and 5 on 3%. Weight those and I give it a .91.

Sea Hag I'd give like 3%, 5%, 7%, 10%, 25%, 50% for a score of 3.99. So sea hag wins.

[HiveMindEmulator]

My criteria is
Value of card X = P(Win) - P(Win | not allowed to buy card X ever)
where the probability is over all possible kingdoms. So a card that can improve your chance of winning by 30% 2% of the time is about as good as a card that can improve you chance of winning by 1% 60% of the time.

This is basically the same as jonts, just with different numbers. Even though no one really does all the math -- since the numbers in our head all so vague that getting more than 2 significant digits out would be impossible -- I think this is generally how people do things at some level.
There is some sense of multiplying the frequency of goodness with the magnitude of goodness.

[Qvist]

So, the poll is closed today and the majority has decided. But I will give the possibility to vote for everything beside of Dark Ages if someone wants to.

[Morgrim7]

when do i submit my list?

[werothegreat]

3.  Because Dark Ages is (likely) never coming to Iso, so it's not like we'll suddenly develop a greater understanding of it later on.

You know, you can play on Goko.

[Qvist]

when do i submit my list?

The polls will probably released at the start of next week.

[ipofanes]

I plan to brazenly rank the Dark Ages cards, despite having only played about 15 games with them.

Let's all theory-rank the Dark Ages cards! That'll be a lark.

In my experience, Mystic is a useless Card. I buy it--and win with it--less than Saboteur. Worst $5 card.

I bet you haven't played Prophecy, the recommended set for Intrigue and Dark Ages.

Armory, Ironmonger, Mystic, Rebuild, Vagrant, Baron, Conspirator, Great Hall, Nobles, Wishing Well

I have. My 9yo pounded me with Rebuild.