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

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.

151

Variants and Fan Cards / Re: "Until your next turn" Attack-Duration

« on: April 30, 2015, 04:04:05 pm »

What if instead of discarding a card, he took his -1 card token each time he played an action?  That seems slightly less powerful.

152

Dominion: Adventures Previews / Re: Blurry gif on Dominion Welt

« on: April 01, 2015, 04:51:30 pm »

Zoom in! Enhance!



The middle card is clearly Gandalf cosplaying as Carmen Sandiego. Does German have a single word for that?

It looks like Wesley Crusher in a Starfleet uniform on the right.  What's the German word for "brat"?

153

Dominion: Adventures Previews / Re: Previews #2 - Guide, Duplicate, Coin of the Realm

« on: March 31, 2015, 10:15:24 am »

Wow, this is really interesting.  I feel like I have no good sense for how often I'd want each of these.

Guide -- Obviously this is going to be good in a game with hand size reducers.  But how often will I want this without those attacks on the board?  It would definitely add reliability to an engine.  If I draw a dud hand, I can just skip over it to my stronger cards.  Will knowing that you have that ability change the way you build engines?

Duplicate -- Kind of like a Smugglers for yourself.  This one's a little trickier than Guide becuase it's not a cantrip, so if you want to do any cool tricks with gaining and playing during the same turn, you're not only going to have to play this early on your turn, you're also going to need to have enough +actions early on to use this and still have the rest of your engine fire.

Coin of the Realm -- This also adds engine reliability.  If I draw terminals together, I just call this back and can play both terminals.  In one sense it compares to Masterpiece since its on-play effect is only a Copper, but the bonus is what you care about.  In another, it's totally different than Masterpiece, since it doesn't spend most of its time siting around in your deck clogging it up.

Overall, reserves are a really cool idea.  I can't wait to use them.  I feel like they're going to change the way I think about Dominion.

154

Dominion: Adventures Previews / Re: Previews #1 - Lost City, Magpie, Hero

« on: March 30, 2015, 01:42:12 pm »

I already like the magpie a lot. The art is very nice and it seems simple to play (no hard decisions to make once bought).
Its a bit comparable to vagrant, but much stronger, as you get another magpie, if you dont hit a treasure. In early game one of the seldom case you might want your card not to hit. The hitting chance is much higher in early game than with vagrant.
So it looks like a decent opener if you expect the game to last long, especially in slogs.
Magpie might also enable engines if there is no copper trashing available.

It looks like there's some synergy between Magpie and a Scrying Pool engine.  Either you draw a treasure into your hand, which helps your Scrying Pool hit more actions or you gain another cantrip, which doesn't really hurt.  Magpie might strengthen Scrying Pool when there's not strong trashing.

155

Rules Questions / Walled Village + Scheme

« on: March 24, 2015, 12:57:09 am »

Apologies if this has been asked before.  A search didn't reveal anything.

Suppose I play three actions during my turn, Scheme, Walled Village, and Smithy.  At the start of my clean-up phase I could presumably choose which of Scheme or Walled Village to resolve first.  If I choose Scheme first, I can put my Smithy on top of my deck.  Now when I resolve Walled Village, my Scheme is the only other action in play.  Am I allowed to put the Walled Village on my deck?  Or is it no longer the start of my clean-up phase?

156

Rules Questions / Re: Embargo tokens on an empty pile

« on: March 14, 2015, 01:48:31 am »

Are you sure you can't do it on Goko?  I'm pretty sure I embargoed an empty pile the other day.

157

Game Reports / Re: What is the fastest way to Kings Court?

« on: March 12, 2015, 01:54:59 pm »

Rats/Watchtower is a really strong opening, also in this kingdom.

On the second shuffle I'd want to get another rats, another watchtower and a worker's village.
After that trashing should go fast. Id you're lucky you can skip silver but getting it is not that much of a problem.
This opening does delay hitting $5 yourself a bit, but on the other hand - I'm not scared at all by the $5 cards with that deck.
Margrave even helps, Ghost ship doesn't hurt as long as you have a watchtower in hand.

Even if this delays KC by a turn or so (and I'm not even sure it does), the trimmed deck should be totally worth it.

Would you play Rats if you happened to draw it with no Watchtower?  (assuming you already have 2)

158

Game Reports / Re: What's the best strategy here? Rats + Death Cart?

« on: March 04, 2015, 12:29:09 pm »

Thanks for the help.  I'm not great at Dominion but I'm trying to get better, and this is really helpful!

159

Game Reports / What's the best strategy here? Rats + Death Cart?

« on: March 02, 2015, 03:15:32 pm »

Code: [Select]

Forager, Death Cart, Scout, Rats, Worker's Village, Counting House, Saboteur, Catacombs, Pillage, Hoard
I was playing this game in real life the other night.  It was a shelters and colonies game, and it looked to me like a Catacombs-BM board, but the trick would be trying to get to Platinum quickly.  I opened Forager+Rats, while my opponent opened Pillage-<nothing>.  My plan was to thin and use Death Cart to vault from Rats to Platinum.  Anyway, the Spoils from Pillage let my opponent get to Golds and Platinums much more quickly.

What do you guys think the winning strategy here is?  Does it involve Rats/Death Cart?  Or just Forager and Catacombs?  Something else?

160

Dominion General Discussion / Re: Dominion Data Mining: Card Rankings

« on: January 28, 2015, 03:42:09 pm »

Thanks for the great work!

I'm quite surprised that Rebuild was only gained in 74% of games.
Did you use all the game logs of the current top 100 players, or only those that were played by a player while they were in the top 100 (which would probably be much more work...)? And did you use a time cut-off (e.g. only games played in the last 12 months) or did you go back till the beginning of Goko?
Either could explain Rebuild's low gain rate (since it was massively underestimated when new), and the first possibility might also explain the high gain rate of the very worst cards.

Thanks.  The data set I used had all games of current top 100 players -- I don't think there's a way on gokosalvager to figure out what their rank was at the time the game was played.  And it goes as far back as the gokosalvager logs go.  I'm not sure if that's all the way back to the start of Goko or not (I would guess that it's not).

161

Dominion General Discussion / Re: Dominion Data Mining: Cards that correlate with skill

« on: January 11, 2015, 03:57:32 pm »

This still doesn't tell me what you did. Yes, I understand bootstrapping - statistics is my job - but I don't know, for instance, what .03 is the standard deviation of. So you took a bootstrap - of what? Games? Players? How big was the bootstrap? How big is the entire thing? You realize that these things aren't uncorrelated, right?

The reason I'm so curious is that 10 standard deviations means actually nothing. Seriously. You make the claim that it means these are really big effects you wouldn't see by random chance. The problem is, if you're implying there's a normal distribution, the chance of getting a result like you claim is so small, I can't get a computer to give me that calculation (at least, within a few minutes on my home machine; point is, it's REALLY small, small enough you need to use a non-standard data type). Even if it's a wacky distribution, Chebyshev's theorem tells us we shouldn't really be getting these kinds of results. So, if I were doing this calculation and getting this result, my natural assumption would be that I had done something wrong in my calculation.

I feel like I get more scrutiny here than with peer review.  The reason I post these is so that people can learn from them if they find them useful.  I have my own thesis I need to finish, so I don't have time to write another one about Dominion.

Normality is a reasonable assumption in the absence of evidence to the contrary, which is why I make that assumption.  I haven't had time yet to see if the distribution of correlation coefficients seems to fit that well.  Your post might be that evidence to the contrary.  The standard deviation of the coefficient averaged over all cards when I run 100 bootstraps on games is 0.01.  The range is 0.007 to 0.032.  I've also done a second experiment where I take each game and ignore which players buy which cards and instead assign these randomly.  This also produces a standard deviation on the correlation coefficient of ~0.01.  In 20+ runs times 206 cards per run, I was not able to get a chance correlation of greater than 0.04.

So it seems very unlikely to me that a card could achieve a correlation of +/- 0.100 simply by chance.

162

Dominion General Discussion / Re: Dominion Data Mining: Cards that correlate with skill

« on: January 11, 2015, 02:08:27 pm »

How do you come up with that?

I used bootstrap sampling.  http://en.wikipedia.org/wiki/Bootstrapping_(statistics)

The highest standard deviations are for Promo cards (which have the least data), so you should be slightly more wary of Promo cards in the original list, but even Prince only has a standard deviation of 0.03.

163

Game Reports / Re: Well...uh...dang.

« on: January 11, 2015, 12:19:03 am »

That's as may be, however I do notice that you've played a couple hundred rated games (over a third of your total) using Pirate Ship trap Kingdoms against humans, with about a 95% win ratio, and usually ending with your opponent having very few VP, suggesting they were successfully pinned.  While I'm sure your opponents gained a new appreciation of the power of Pirate Ship, I personally have never found pin traps to be particularly fun (from either side).  If you want to make this board a bit more sporting, I would suggest adding Secret Chamber, so that there is at least a chance of being able to escape the trap.

I think a really fun way out of the trap would be to add Masquerade.

164

Dominion General Discussion / Re: Dominion Data Mining: Cards that correlate with skill

« on: January 10, 2015, 11:44:27 pm »

I have a suggestion for another similar analysis.  (this one is definitely cool)

If only one player in a game buys a card, and the other player does not, what % of the time is the player who bought the card the one with a high skill rating? (You could weight it by the skill gap, but I don't think that would be that helpful)

Definitely.  That's one of the things I'm planning to do next with this data.

Can someone estimate the likelyhood of a card getting +/-0.100 correlation just by chance? Or do we need further assumptions? We have more than 200 cards and I'm wondering whether some outliers could be explained by this.

A correlation of +/- 0.100 would be about 10 standard deviations from the mean, so it's pretty safe to say that none of the results in the first post are due to chance.

I must've missed the previous thread... has someone put up that whole DB to be downloaded?  I have an experiment or two I'd like to try myself. 

It's actually quite easy to get this data.  The data is from gokosalvager.com.  All I did was download all the game logs for each of the top 100 players.

165

Dominion General Discussion / Re: Dominion Data Mining: Cards that correlate with skill

« on: January 09, 2015, 04:53:50 pm »

(does anyone know why people talk about explaining X% of the variance and not the deviation? why is [unit^2] more meaningful than [unit]?)

Probably the main reason that we talk about variance rather than deviation is that variance is always positive (since it's a squared value), whereas deviation can be positive or negative.  So you'd have to use the absolute value of the deviation, which gets messy mathematically.  There's also a nice linear relationship with sum of squared deviations that isn't there when you try to sum absolute deviations (but that's more on the technical side).

is it possible to represent these data as some kind of function like, difference in likelihood of purchase/number purchased as a functino of difference in rank?  These correlations seem hard to interpret.

To get such a function, I would need to run linear regression on this data and pull out those coefficients.  The correlation coefficient (what's presented in the first post) simply measures how close to linear the relationship between skill and gaining a certain card is.  It ignores the strength of that relationship.

166

Dominion General Discussion / Re: Dominion Data Mining: Cards that correlate with skill

« on: January 09, 2015, 11:02:37 am »

So, the biggest thing to note here is that all of the numbers are tiny. You're not finding anything significant. Well, maybe statistically significant, but not practically so.

I'd argue that finding correlations this large is actually fairly substantial.  If we assume that how often you buy a card is independent of other cards (which is a fairly reasonable assumption as far as independence assumptions go, since in full random the chance of getting any two specific cards in a kingdom is ~0.002%), then the r^2 values range from 0.7% to 4%.  This means that statistically, I can explain 4% of the variation in skill among players simply by looking at how often the player buys Governor.  If you sum up the top 20 cards on the weighted list, that explains 29% of the variance in the skill.

That's huge.  This doesn't even include things like how cards are played once they're bought, when to start greening, etc.  So the fact that we can explain so much of the variance in skill simply by a how often a few cards are bought is a really big deal.

167

Dominion General Discussion / Re: Dominion Data Mining: Cards that correlate with skill

« on: January 09, 2015, 12:13:42 am »

The main problem is that you are trying to estimate an association between two binary variables (high-ranked / low-ranked with buy/don't buy).
...

Thanks for the good thought Merudo.  Fortunately, I think the problem is not quite that dire.  Player skill is actually a continuous variable, which makes the correlation much more reliable.  That being said, the skill distribution in my dataset is biased by my collection method.  Just so that everyone is aware of what kind of biases might be playing into these results, I've attached a histogram of the skill levels represented.

168

Dominion General Discussion / Dominion Data Mining: Cards that correlate with skill

« on: January 08, 2015, 12:52:17 pm »

Working off suggestions from the previous thread, I've taken the 90,000 game database of games by top-100 players and looked at how the player's skill (represented by their TrueSkill rating) correlates with what cards they gain.  Fortunately, top-100 players also play plenty of low-ranked players, so there are lots of examples of both low- and high-ranked players in the database.

The table below shows two different rankings.  Both measure the correlation coefficient between gaining a certain card and the skill of the player.  Positive numbers mean that the card tends to be bought more often by high-ranked players than low-ranked players.  Negative numbers mean that the card tends to be bought more often by low-ranked players.  The ranking on the left is an unweighted ranking, which means that the correlation is between skill and whether the card gets gained or not (gaining it multiple times doesn't make any difference.  The ranking on the right is a weighted ranking, which means that here the number of times the card is gained makes a difference.    So on the right side, you can say, for example, that good players seem to know to get lots of wharves, whereas bad players seem to overbuy some good cards like Tournament or Bishop.

	Unweighted ranking		_____________________		Weighted ranking
Rank	Card	Correlation		Rank	Card	Correlation
1	Butcher	0.127		1	Governor	0.206
2	JackOfAllTrades	0.106		2	Wharf	0.172
3	Wishing Well	0.104		3	JackOfAllTrades	0.148
4	Vineyard	0.087		4	Hunting Party	0.144
5	Governor	0.085		5	Wishing Well	0.141
6	Chancellor	0.081		6	Vineyard	0.138
7	Duke	0.080		7	Stonemason	0.112
8	Courtyard	0.077		8	King's Court	0.111
9	Warehouse	0.071		9	Apothecary	0.111
10	Masterpiece	0.070		10	Butcher	0.107
11	Scavenger	0.065		11	Duke	0.106
12	Fairgrounds	0.062		12	Stables	0.100
13	Oracle	0.056		13	Scrying Pool	0.097
14	Journeyman	0.056		14	Horn of Plenty	0.096
15	Apothecary	0.055		15	Fairgrounds	0.095
16	Masquerade	0.055		16	Menagerie	0.095
17	Counterfeit	0.054		17	Oracle	0.094
18	Duchess	0.051		18	Watchtower	0.093
19	Stonemason	0.050		19	Masquerade	0.088
20	Ambassador	0.041		20	Warehouse	0.087
	...				...
187	Moneylender	-0.094		187	Spy	-0.089
188	Trade Route	-0.096		188	Remodel	-0.093
189	Expand	-0.096		189	Forge	-0.096
190	Treasure Map	-0.099		190	Soothsayer	-0.102
191	Coppersmith	-0.103		191	Tournament	-0.102
192	Marauder	-0.106		192	Island	-0.105
193	Soothsayer	-0.107		193	Expand	-0.105
194	Talisman	-0.113		194	Baker	-0.106
195	Transmute	-0.115		195	Feast	-0.106
196	Spy	-0.118		196	Trade Route	-0.107
197	Golem	-0.118		197	Transmute	-0.109
198	Feast	-0.123		198	Tribute	-0.114
199	Mine	-0.128		199	Bishop	-0.119
200	Alchemist	-0.132		200	Treasure Map	-0.127
201	Tribute	-0.155		201	Mine	-0.130
202	Bishop	-0.158		202	Scout	-0.145
203	Scout	-0.171		203	Marauder	-0.146
204	Taxman	-0.177		204	Saboteur	-0.168
205	Saboteur	-0.195		205	Taxman	-0.172
206	Pirate Ship	-0.262		206	Pirate Ship	-0.204

169

Dominion General Discussion / Re: Dominion Data Mining: Card Rankings

« on: January 08, 2015, 12:56:52 am »

Could you generate same statistic for people with different levels to see what weaker players over/underrate?

I actually have this data calculated and ready to report.  Would the mods prefer that I post it in this topic or create a new topic for it?

170

Variants and Fan Cards / Re: Sage for bad cards

« on: January 07, 2015, 12:30:34 am »

You discard all the revealed cards? If so, then this is really just doing blueblimp's venture thing. You should probably put the junk card in hand. Then there's some combos with Cellar or Crossroads or something.

Agreed.  If the junk card went in hand, this would be a good pickup in the early game as it would often draw a copper.

171

Dominion General Discussion / Re: Dominion Data Mining: Card Rankings

« on: January 05, 2015, 10:06:44 pm »

for games with 2 top 100 players, did you count both?

Yes, if there are two top-100 players, then each kingdom card is counted as being available once for each player.

How many different kingdom cards are bought on average? Looks like 5 or 6. Also looks like only about say 15% of boards are BM boards, given the frequency that cards that are always bad in BM are bought.

Not sure if this is exactly what you're asking, but 9.85 differently named cards (modulo me grouping the Ruins and Knights together) are gained per player per game.

Do you have any data about win rates with/without gaining specific cards? This would suggest which cards are under/over rated.

Maybe I'll do this next.  It should be pretty straightfoward.

Is it possible to distinguish between Province and Colony games in the data?

Yeah, I can just look at whether Colony is in the supply.  What did you have in mind?

172

Dominion General Discussion / Re: Dominion Data Mining: Card Rankings

« on: January 05, 2015, 09:38:12 pm »

I wouldn't be surprised if Swindler alone accounts for as much as half of the Adventurer gains. Add Colony games, Fool's Gold, Horn of Plenty and Fairgrounds to your consideration, and Adventurer's popularity should be a bit less surprising.

Since you made me curious, here are the top 20 forced gains from Swindler.  Surprisingly this explains at most about 10% of games where a top 100 player gains Adventurer.

1	Curse	5357
2	Estate	3115
3	Silver	1992
4	Duchy	1584
5	Gold	1383
6	Province	849
7	Copper	763
8	Swindler	591
9	Potion	467
10	Chancellor	164
11	Masterpiece	148
12	Loan	134
13	Doctor	130
14	Woodcutter	118
15	Lookout	115
16	Ruins	103
17	Develop	101
18	Platinum	95
19	King's Court	94
20	Mine	93
	...
31	Adventurer	67

173

Dominion General Discussion / Re: Dominion Data Mining: Card Rankings

« on: January 05, 2015, 05:05:46 pm »

Do you distinguish between voluntary gain and forced gain (i.e., Swindler)?  Doubt it'll have much impact but mostly just curious.  Throwing in the base cards (Copper, Curse, Estate, etc.) might be interesting too.

I don't currently have a way of differentiating those two simply from the information in the logs, especially because sometimes the forced gain is desired, a la Border Village or Catacombs.  But I expect that the effect is quite minor, especially since the majority of forced gains in attacks are with base cards.

174

Dominion General Discussion / Prizes

« on: January 05, 2015, 04:40:40 pm »

Prizes

Rank	Card	+/- vs. Qvist	% games gained	# games gained	# games available
1	Followers	0	43.1%	3372	7825
2	Trusty Steed	0	43.1%	3369	7825
3	Princess	0	30.7%	2402	7825
4	Bag of Gold	0	18.9%	1479	7825
5	Diadem	0	15.8%	1240	7825

175

Dominion General Discussion / Potion cost cards

« on: January 05, 2015, 04:40:14 pm »

Potion cost cards

Rank	Card	+/- vs. Qvist	% games gained	# games gained	# games available
1	Scrying Pool	0	67.3%	5142	7634
2	University	+4	64.9%	4919	7574
3	Familiar	-1	60.0%	4553	7592
4	Alchemist	0	52.9%	3986	7534
5	Vineyard	-2	47.2%	3578	7588
6	Apothecary	-1	47.1%	3570	7575
7	Golem	0	40.4%	3032	7503
8	Possession	0	37.8%	2815	7452
9	Philosopher's Stone	0	12.1%	907	7518
10	Transmute	0	7.6%	584	7661