Print

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

[markusin]

Banlist:
Possession
Urchin
Peddler
Lurker
Swindler
---
Defiled Shrine
Tax
Keep

Possession: I've seen enough degenerate stuff with this card to not be thrilled about seeing it anytime soon, and this includes both players "debt locking" themselves. For every good Possession game, there are at least two others that will feel dumb.

Urchin: the disparity that can arise despite both players opening the same way is absurd, and I am never happy about being called a "lucker" for luck being on my side with it.

Peddler: player A gets 3 peddlers, then player B gets 3 peddlers, then player A gets another 4 peddlers and wins the split 7-3. This is the most likely sort of outcome in any game where Peddler actually matters. Bonus points if Remodel or similar is on the board.

Lurker: the cool pileouts are not worth the standoff where neither player plays their lone Lurker. Honestly though, this card is a headache for me to use properly.

Swindler: Having your opening $5 turned into a Duchy (or Goat to Estate) feels like too much of a handicap in many cases. Where are my Goko zaps? Bonus points if the $5 swindled on the second shuffle was the Treasury you opened with (true story).

Defiled Shrine: it goes from bad buy to great within the span of a turn start to buy phase and one player gets first dibs at that tipping point, but also the setup is messy.

Tax: A messier setup than Defiled Shrine and messes with openings in an annoying, non-symmetric way.

Keep: Counting Coppers? I'll leave that to the misers and those guys who settled at the house of counting. Having to win the Contraband split is funny for only so long.

Dislikes:
Cultist
Margrave
Scrying Pool
Ghost Ship
Advisor
----
Donate
Save
Fleet

Regarding the vertical cards here, they either have too much snowball potential that is not very well offset by the games where they are interesting, or they involve too much clicking (and some have both, you know who you are). Ambassador gets a pass because often times both players can get thin, and there there is an interesting balancing act to play.

Donate, it's too much of a headache sometimes, and it can heavily favour first player. Save, it's too automatic a decision to get It, and then you often have to avoid hitting "play all treasures" and it's a bit annoying to worry about that every turn. Fleet, uh, I'm not comfortable with it yet and it's kind of a downer card, either creating imbalances or just making things way too mathy. I'll probably fill this with something else in the future.

Likes:
Stonemason
Castles
Pixie
Procession
Border Village
----
Innovation
Ball
Orchard

Stonemason: Actions galore, "true" splitter, out of nowhere pileouts. Nice!

Castles: it's funky alt-VP, without making things too sloggy.

Pixie: This card leads to lots of interesting decisions, and risk/reward assessments, with Goat making life easier all around.

Procession: I like fireworks. Procession delivers.

Border Village Lots of actions. Yay! Not too swingy, as it mainly shows its strength when multiples are gained.

Innovation: It feels so good to meld the first card I gain in my turn and get a free activation of its dogma effect.

Ball: Perfectly ballanced, as all things should be.

Orchard: Orchard encourages fulfilling the Dominion dream - getting lots of stuff, and gettings lots of that stuff. Any hall of oddities carny can get a single copy of some throwaway card on the last few turns of a game, but Orchard involves more finesse. You need to commit to those Mines and Trading Posts if you want Orchard points off them.

Thank you for reading.

[Donald X.]

People get matched against people of similar skill levels; the skill multiplier isn't the story of good players beating bad players. Black Market has a skill multiplier of 1.23, which is crazy high. It feels high-luck; it's actually high-skill. As further evidence, it's Stef's favorite card. Smithy meanwhile is high-luck and no-one complains about it. That's just how it goes. (Swindler is an example of a card that looks high-luck and is).

You can choose the level of the players you want to be matched with, right? I have +/- 10, so I'm usually matched with people 8-10 levels lower than me. That usually means I'm a better player. Still, I have very rarely seen an opponent choose the wrong cards from the BM deck. If a junker is missing, they buy the junker. If a village is missing, they buy the village. It's not that complicated. Maybe with very high-level play, the opportunity to construct a crazy engine via BM makes it a high-skilled card in that sense, but in most games where one of the players has my skill level (i.e. me), it's how I described. That's been my experience, which is all the evidence I have. Whether it's Stef's favorite card doesn't really say much, sorry.

Stef is the highest rated player. That doesn't prove his favorite card wouldn't be a luck-fest; but, the odds are against it.

The philosophy, "x is missing, buy x" sounds like a poor way to do well with Black Market. What does your deck need now, is where to start. "!stats black market buys" on the discord will get you a chart of how Black Market buys correlate with winning (for a small number of games, minimum of 50). You didn't give "trasher" as an example, but trashers from the Black Market tend to suck. That's not obvious to everyone, as we can see from the people who bought a Black Market Chapel anyway. If a junker is missing, they like to buy the junker; Mountebank is one of the most-gained cards on the list. It's not at the top in terms of how it performs for you though (though it's fine), just for how often people take it.

If a village is missing, Black Market is doing way less, may not be worth having.

I also really wonder what exactly it means that Smithy is high-luck? Seems odd. Maybe my definition of high-luck and whatever the skill multiplier signifies is different.

The skill multiplier means that our ability to predict the outcome of a game with that card, based on the ratings of the players, is that much affected. There's a video by aku chi that explains all of the stats at length; it was way long and I haven't sat through it. markus made the charts and well on the discord you can ask questions about it in #skynet.

It seems clear that there are luck-based scenarios for Smithy, e.g. do you draw it with a village or not, that could affect its skill multiplier. However Laboratory also has a low skill multiplier. So probably a bigger factor is, these cards are so simple. They don't give you anything to finagle well.

[Jeebus]

Stef is the highest rated player. That doesn't prove his favorite card wouldn't be a luck-fest; but, the odds are against it.

BM is my least favorite card, so going by this metric, I should be among the lowest rated players. But I'm among the highest. Just like Stef, I'm exactly one person. As I said, this doesn't say much.

The philosophy, "x is missing, buy x" sounds like a poor way to do well with Black Market. What does your deck need now, is where to start.

I meant that if an engine can be built but something is missing (engine pieces or payload), BM is usually worth going for, because if you get what you need, obviously the earlier the better, you will probably win (given that the other player is less lucky with BM). Yes, obviously that missing piece might not be worth buying if it's too late. Of course you always look at what your deck needs.

The skill multiplier means that our ability to predict the outcome of a game with that card, based on the ratings of the players, is that much affected. There's a video by aku chi that explains all of the stats at length; it was way long and I haven't sat through it. markus made the charts and well on the discord you can ask questions about it in #skynet.

It seems clear that there are luck-based scenarios for Smithy, e.g. do you draw it with a village or not, that could affect its skill multiplier. However Laboratory also has a low skill multiplier. So probably a bigger factor is, these cards are so simple. They don't give you anything to finagle well.

I have not tried to understand how the skill multiplier is calculated, nor what that actually means in practice (which are two different things) - and I won't. I think it's not very unlikely that it doesn't actually mean what you (not just meaning you Donald) think it means. At the very least it's clear that, even if we assume that a high score identifies a "high-skill card", the opposite of such a card is not a "high-luck card", since it's absurd to say that Laboratory (or even Smithy) is high-luck.

[Donald X.]

BM is my least favorite card, so going by this metric, I should be among the lowest rated players.

No, there's no sense to what you're saying. If you want me to take what you say seriously, I don't understand why you would say something like this. I'll guess that you don't. That's a bummer, since you're that rules guy.

I have not tried to understand how the skill multiplier is calculated, nor what that actually means in practice (which are two different things) - and I won't. I think it's not very unlikely that it doesn't actually mean what you (not just meaning you Donald) think it means. At the very least it's clear that, even if we assume that a high score identifies a "high-skill card", the opposite of such a card is not a "high-luck card", since it's absurd to say that Laboratory (or even Smithy) is high-luck.

The presence of Laboratory in a game reduces the chance that the higher-ranked player wins. I don't need to argue over what we call that.

It sounded like you were interested, but man no-one needs you to understand or care about it. It's real data; I trust it way more than gut feelings.

People are incredible at believing false things. At this point I'd tell you to look up cognitive biases, but I mean, let's guess that you don't care about those either. Most people don't. I always think of how Paul Erdos couldn't believe the answer to the Monty Hall problem. He didn't accept it until seeing a computer simulation. How is his name on so many math papers, we might wonder.

No-one needs you to understand or like Black Market. It's great that you can not play with it, as in real life.

[Jeebus]

BM is my least favorite card, so going by this metric, I should be among the lowest rated players.

No, there's no sense to what you're saying. If you want me to take what you say seriously, I don't understand why you would say something like this. I'll guess that you don't. That's a bummer, since you're that rules guy.

I am taking you seriously. That's why I'm replying directly to what you wrote. Even though you didn't, I'll explain my point. I'm saying that it doesn't matter much what one person views as his favorite card. If Stef's being a good player and having that as a favorite card is evidence that the card is high skill, why isn't my being a good player and having it as my least favorite card evidence to the contrary? Of course I'm not as good as him, but I'm still among the best players. This tells us that there is something lacking with taking this as evidence.

I have not tried to understand how the skill multiplier is calculated, nor what that actually means in practice (which are two different things) - and I won't. I think it's not very unlikely that it doesn't actually mean what you (not just meaning you Donald) think it means. At the very least it's clear that, even if we assume that a high score identifies a "high-skill card", the opposite of such a card is not a "high-luck card", since it's absurd to say that Laboratory (or even Smithy) is high-luck.

The presence of Laboratory in a game reduces the chance that the higher-ranked player wins. I don't need to argue over what we call that.

It sounded like you were interested, but man no-one needs you to understand or care about it. It's real data; I trust it way more than gut feelings.

Sure, but data can be interpreted in many ways.

We were both talking about whether BM was a high-luck card. You used the term high-luck as the opposite of high-skill. Maybe "the presence of Laboratory in a game reduces the chance that the higher-ranked player wins" doesn't mean that Laboratory is high-luck. Maybe "the presence of Black Market in a game increases the chance that the higher-ranked player wins" doesn't mean that Black Market is not high-luck. In fact there was a thread about BM where someone was saying that a card could be both high luck and high skill.

We might also be using our cognitive biases when we interpret data.

[DDL]

Yes it says it decreases the probability of the card being chosen.

But on my first 3 plays after this my two disliked cards both turned up once.

They were
Sauna as it seems to be the one way (althoughif only one person goes for it it might be OK
Feodum as I can't see itr being useful. *although I did have a game with masterpiece and other players did not understand the cards so I emptied the silver pile before buying any

Trader + Feodum is crazy good. But it's hard to get both cards in the same kingdom, statistically.

[Donald X.]

I'm saying that it doesn't matter much what one person views as his favorite card.

It does not matter what one person views as their favorite card.

Like any argument-by-authority, it was an attempt to get somewhere faster than having a long real argument. I also gave a real argument, but then, maybe this argument by authority would do something for you, is what I thought.

I am confident that Stef views Black Market as high skill (in addition to it being his favorite card). But, I didn't want to go ask him, to have up-to-the-minute data, and improve what would still be an argument by authority. He does cite it as tricky to play well in his article on it.

If Stef's being a good player and having that as a favorite card is evidence that the card is high skill, why isn't my being a good player and having it as my least favorite card evidence to the contrary? Of course I'm not as good as him, but I'm still among the best players. This tells us that there is something lacking with taking this as evidence.

I was not factoring in any skill level on your part; that is not information I had before, and all I know now is that you say you're good, though I can believe you're at whatever level you say. Stef is good; I know that. If I had started out thinking e.g. "Jeebus is basically 2nd best to Stef" then I would not have offered up, in addition to my real argument, that argument by authority.

So far my estimate of how good you are goes down due to seeing that you think Black Market is low-skill. But, that could just be your weak point; the leaderboard tells a better story.

Sure, but data can be interpreted in many ways.

We were both talking about whether BM was a high-luck card. You used the term high-luck as the opposite of high-skill. Maybe "the presence of Laboratory in a game reduces the chance that the higher-ranked player wins" doesn't mean that Laboratory is high-luck. Maybe "the presence of Black Market in a game increases the chance that the higher-ranked player wins" doesn't mean that Black Market is not high-luck. In fact there was a thread about BM where someone was saying that a card could be both high luck and high skill.

We might also be using our cognitive biases when we interpret data.

Luck is not the opposite of skill; they are two different axes. Poker is high skill high luck, as is Dominion; Chess is high skill low luck; Tic-Tac-Toe low skill low luck, Chutes and Ladders low skill high luck.

But, making the better player lose more often is just what people in general, but who knows about present company, would call "high luck." In conversations about cards being swingy, it is rare that anyone is interested in getting into the fine details of luck and skill. And your initial post called Black Market out as being low skill, not just high luck.

Finally get to ban Black Market. You can't convince me that this isn't the most swingy card in the game. Maybe it's high-skill in that it's a good card to beat bad players with, but when both players are decent, it's just "who get's the missing engine pieces, trashers or junkers - and who doesn't? let's roll the die".

If that is not you saying that Black Market is low skill, then we have a significant language barrier.

We are all affected by cognitive biases constantly; there's even a cognitive bias that makes you think you're less affected by cognitive biases due to knowing about them. One thing I do is ask disinterested third parties what they think about situations where I'm trying to rule out bias on my part.

[DDL]

This article sold me on BM:

http://forum.dominionstrategy.com/index.php?topic=11896.msg428733#msg428733

[Jeebus]

But, making the better player lose more often is just what people in general, but who knows about present company, would call "high luck." In conversations about cards being swingy, it is rare that anyone is interested in getting into the fine details of luck and skill. And your initial post called Black Market out as being low skill, not just high luck.

It's just that you called Smithy (and then Lab) high-luck. I saw this as a weakness in the data, or rather the interpretation of it. Like you, and most people, I was confounding high-luck and low-skill (and vice versa) when I wrote what you quoted from me about BM. I do think that it's high-luck, and I do think that it's not particularly high-skill, but as we have learnt those are not necessarily related. Smithy/Lab is certainly neither high-luck or high-skill.

But later I was mainly talking about the data and why I don't really trust the way people normally use it. I was saying that even if we assume that the data is correct in how it flags high-skill cards (which would mean I'm wrong about that aspect of BM), the data doesn't seem to say much about a card's luck level. Maybe we agree on that part?

But regarding the data in general. I don't really have that great of an impression about how accurate it is (meaning how it's interpreted - I assume the data itself is correct). Maybe if I took the time to delve deep and research it thoroughly, I would be converted, but I want to use my time... arguing in the forum instead. :p Isn't it the same data being used in the "best card" threads to talk about how good a card is based on how likely buying it makes you win? There have been some crazy claims based on that data that nobody seemed to understand.

[Jeebus]

This article sold me on BM:

http://forum.dominionstrategy.com/index.php?topic=11896.msg428733#msg428733

Interestingly, Stef said there that a trasher from the BM deck is a "bomb" (an "exceptionally strong card") - which matches my experience, of course if a strong trasher is missing from the kingdom. But Donald said they tend to suck, based on the data I assume. Is Stef wrong or the data?

[elahrairah13]

People get matched against people of similar skill levels; the skill multiplier isn't the story of good players beating bad players. Black Market has a skill multiplier of 1.23, which is crazy high. It feels high-luck; it's actually high-skill. ... Smithy meanwhile is high-luck and no-one complains about it. That's just how it goes. (Swindler is an example of a card that looks high-luck and is).
...
The hexes have a skill multiplier of 1.00; they don't have an effect on our ability to predict the winner from skill level. Isn't that interesting.

I'm super interested in this data, can a kind soul direct me to the best way to find it?

[markusin]

This article sold me on BM:

http://forum.dominionstrategy.com/index.php?topic=11896.msg428733#msg428733

Interestingly, Stef said there that a trasher from the BM deck is a "bomb" (an "exceptionally strong card") - which matches my experience, of course if a strong trasher is missing from the kingdom. But Donald said they tend to suck, based on the data I assume. Is Stef wrong or the data?

Many of the "tempo" trashers seem to do well off the BM. This includes Butcher, Junk Dealer, Sentry, Masquerade, Counterfeit, and Recruiter, among others. The trashers that don't do so well include Forager, Remake, Sacrifice, Spice Merchant, Raze. It's an odd pattern.

[Jeebus]

Many of the "tempo" trashers seem to do well off the BM. This includes Butcher, Junk Dealer, Sentry, Masquerade, Counterfeit, and Recruiter, among others. The trashers that don't do so well include Forager, Remake, Sacrifice, Spice Merchant, Raze. It's an odd pattern.

I noticed that all the trashers that do well, except Masq, cost $5. The ones that don't do well, cost less. I have no idea what that means. I was going to say opportunity cost, but then I would expect it to be the reverse.

[MiX]

Many of the "tempo" trashers seem to do well off the BM. This includes Butcher, Junk Dealer, Sentry, Masquerade, Counterfeit, and Recruiter, among others. The trashers that don't do so well include Forager, Remake, Sacrifice, Spice Merchant, Raze. It's an odd pattern.

I noticed that all the trashers that do well, except Masq, cost $5. The ones that don't do well, cost less. I have no idea what that means. I was going to say opportunity cost, but then I would expect it to be the reverse.

I suppose you have $5 when you draw BM? Is that the missing link, that you can pretty much always afford the 5 cost trashers, and since they're stronger, you win more with them?

[Commodore Chuckles]

Interestingly, Stef said there that a trasher from the BM deck is a "bomb" (an "exceptionally strong card")

I don't consider Bomb to be exceptionally strong, but it does strongly effect gameplay, particularly in a destructive way.

[markus]

Skill multiplier is not a very difficult concept: you are say 1 mu (roughly 7.5 levels) ahead of your opponent on the leaderboard. This has some implications for how much you win on an average board (73%). A skill multiplier of 0.8 means that with that card on the board, you only win as if the skill difference was 0.8*1 (resulting in 69%).
"The better player winning less than the worse player" would only happen if the skill multiplier was negative - that is by far not the case for any card, and I doubt that such a card will ever exist (e.g. "The weaker player tosses a coin and wins the game on heads.").

A high skill multiplier might mean that it's difficult to master the card, or it involves little luck such that small skill differences really matter, or that it's better known to good players (e.g. some players might know the base cards but not the expansions). A single number can't really tell apart 3 reasons, but for Black Market the total effect is that the better player wins more than expected. Hence, even if there was more "luck" involved, there is even more "skill" needed.

[Cave-o-sapien]

I always think of how Paul Erdos couldn't believe the answer to the Monty Hall problem. He didn't accept it until seeing a computer simulation.

This is the most interesting thing I've read all day. How did I not know this before?

[JW]

A high skill multiplier might mean that it's difficult to master the card, or it involves little luck such that small skill differences really matter, or that it's better known to good players (e.g. some players might know the base cards but not the expansions). A single number can't really tell apart 3 reasons, but for Black Market the total effect is that the better player wins more than expected. Hence, even if there was more "luck" involved, there is even more "skill" needed.

To add another possibility: a card isn't that good (or it's fine to get one, but you shouldn't buy a lot), but it takes a more skilled player to recognize that.  For example, less skilled players often overvalue coffers, and therefore go for mass Candlestick Maker strategies (which are usually terrible), and mass Baker strategies (which are often sub-optimal).

[David Goldfarb]

Liked:
Captain and Church, since they are new and I want to see how they impact the game. (Despite being on the list, I still haven't had a game with them.)

Disliked:
Ill-Gotten Gains and Governor warp the game too much towards one particular strategy.

Cultist is too swingy: if you and I both go for Cultist and manage to buy two of them, and you luck into a chain where you play them both on the same turn while I don't, you now have a huge advantage. 7-3 splits on the Ruins are quite common, and too hard to come back from.

Knights: Just annoying.

Wall: Likewise.

Banned:
Rebuild: Same objection as IGG and Governor, but even more so. As others above have noted, Rebuild lets you skip building your deck: just get $5 a couple of times and go straight to greening. Since the fun of the game is in building your deck, why would I want to play this game?

Scrying Pool: leads to games with too much downtime between your turns.

Keep: Someone upthread said this warps the game too much, and I agree. (He actually prompted me to add this to my list just now.)

[LibraryAdventurer]

Banned:
Possession
Tournament
Sea Hag
Lurker

Disliked:
King's Court
Page
Ill Gotten Gains
Pillage
Taxman
Keep, Tax, Salt the Earth

Liked:
Black Market
Throne Room
Scepter
Save, Piazza, Star Chart
and Captain and Church just to see what they're like

[Accatitippi]

The skill multiplier means that our ability to predict the outcome of a game with that card, based on the ratings of the players, is that much affected. There's a video by aku chi that explains all of the stats at length; it was way long and I haven't sat through it. markus made the charts and well on the discord you can ask questions about it in #skynet.

It seems clear that there are luck-based scenarios for Smithy, e.g. do you draw it with a village or not, that could affect its skill multiplier. However Laboratory also has a low skill multiplier. So probably a bigger factor is, these cards are so simple. They don't give you anything to finagle well.

Yeah, considering that (if I recall correctly) those stats are built on the top percentile or so of players, we probably can assume that the effectiveness of simple cards like those has already plateaued at top levels. That said, non-elite players have a lot to finagle with there.

The skill multiplier really proxies the question: what cards are top top players using more effectively than slightly-less-top top players?
Looking at it from a different point of view, a card having a low skill multiplier means that to become a top player you must play that card almost as well as currently possible. A higher skill multiplier, conversely, indicates that perfecting your use of that card is important to rise to the top of the top.

Anectodally, I'm sure that that one time I played with Stef, his handling -my mishandling- of Wharf (Skill Multiplier 1.00) was determinant to me being trounced.

[Donald X.]

It's just that you called Smithy (and then Lab) high-luck. I saw this as a weakness in the data, or rather the interpretation of it. Like you, and most people, I was confounding high-luck and low-skill (and vice versa) when I wrote what you quoted from me about BM. I do think that it's high-luck, and I do think that it's not particularly high-skill, but as we have learnt those are not necessarily related. Smithy/Lab is certainly neither high-luck or high-skill.

You have one card instead of having another. The presence of Laboratory in a kingdom means there's one less slot that could be something that favored the better player. However innocent it looks, it being in a game, again instead of something else, reduces our ability to predict the winner based on ranking.

But later I was mainly talking about the data and why I don't really trust the way people normally use it. I was saying that even if we assume that the data is correct in how it flags high-skill cards (which would mean I'm wrong about that aspect of BM), the data doesn't seem to say much about a card's luck level. Maybe we agree on that part?

Skill multiplier is simply the card's effect on our ability to predict the winner based on ranking. It can't separate out luck vs. skill. Randomness on the part of the card will make us less able to predict the winner based on ranking; trickiness will make us more able to.

There's lots of other data to look at too, to try to understand the impact of the card.

But regarding the data in general. I don't really have that great of an impression about how accurate it is (meaning how it's interpreted - I assume the data itself is correct). Maybe if I took the time to delve deep and research it thoroughly, I would be converted, but I want to use my time... arguing in the forum instead. :p Isn't it the same data being used in the "best card" threads to talk about how good a card is based on how likely buying it makes you win? There have been some crazy claims based on that data that nobody seemed to understand.

When aku chi types "!stats lurker" and an image with lots of stats comes up, yes, it's that data. You can also see some of it in chart form covering multiple cards. markus wrote the software that assembles the charts.

The stats are a lesson in how tricky statistics can be. For example, one thing we saw early on was cards where it was good to have 2 or 4 or 6 more copies than your opponent, but bad to have 3 or 5 more copies than your opponent (good/bad meaning, how often does a player in that situation win). We realized that this was because if you have 2 or 4 or 6 more copies, maybe the pile emptied, but if you have 3 or 5 for, it didn't (except in exotic situations). And markus added a line for when the pile is empty so you could see that on the chart.

[Donald X.]

I'm super interested in this data, can a kind soul direct me to the best way to find it?

On the Dominion discord, you can type !help to get a list of dombot commands, and one of them is !stats. With e.g. "!stats fishing village" you get a page of stats for fishing village. There are also some special pages you can get, e.g. trashers, black market buys. And you can private message dombot if you don't want to share the stats with whatever channel.

A magical link: https://discord.gg/jNTnqm

[Donald X.]

Interestingly, Stef said there that a trasher from the BM deck is a "bomb" (an "exceptionally strong card") - which matches my experience, of course if a strong trasher is missing from the kingdom. But Donald said they tend to suck, based on the data I assume. Is Stef wrong or the data?

Many of the "tempo" trashers seem to do well off the BM. This includes Butcher, Junk Dealer, Sentry, Masquerade, Counterfeit, and Recruiter, among others. The trashers that don't do so well include Forager, Remake, Sacrifice, Spice Merchant, Raze. It's an odd pattern.

I think the key thing is that you get Black Market cards later. You don't get them turns 1-2; you only have one shot at getting them turns 3-4 (well two shots if you open with two Black Markets). Many of the trashers just aren't good enough (typically) if you don't get them right away (and people get them from the Black Market anyway).

The data for Black Market buys is scant though, 50-150 games per card.

[apostolosoruler]

i think black market is indeed a very high skill and high luck card, but the skill is the main thing there. the decisions related to if you buy it, when you buy it, why you buy it, if you play it compared to another terminal if there is a collision and finally and most importantly, what you choose or you dont choose from it are extremely skill dependent and completely different on each kingdom. that being said, i dont think an article from 5 years ago, even if it is from stef which means its top notch, can be taken into full account. the evolution of dominion and i can only account for the last year and take the previous years into consideration only through articles etc, is huge and many major changes in thinking have taken place. therefore, taking into consideration the time of sth should allways be the start. finally, i would like to add that you very often see even the very top players do mistake with what they choose or dont choose from black market, so you know that is a place where a lot of improvement can be made.