Print

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

[SirPeebles]

Hey all.

What are some interesting strategies you've seen bots follow?  For the most part, I'm impressed.  Obviously the AI isn't perfect, but they generally tend to follow actual, well-known strategies.  Sometimes they'll go for big money + wharf. Sometimes they'll fight to buy up all of the minions.

I just played a game where the bot built a golden deck.  Unfortunately, the golden deck was two bishops and a workers village.  Each turn he trashed two coppers and bought two more, netting only two points.  He could have at least bought an estate and copper, netting three points per turn.

I've also noticed bots seeming to ignore embargoes.  Earlier today I opened 5/2 and decided to embargo gold while using bandit camps for spoils.  The bot seemed unphased and bought a gold anyhow.  I dropped a second embargo on gold and he still bought another.  I even dropped a third embargo, and he bought yet another gold.  A total of three golds accompanied by six curses.

[ftl]

The bots are terrible with trash-for-benefit in general; they trash the wrong things at the wrong times. With Bishop they can end up trashing their whole deck like that. I've seen the bots make some brutal chapeled-down engines though.

[DG]

The bots tend to have some good strategies based on cards pairings and some poor strategies when things get more complicated. Some bots do recognize the embargo and this led to the only occasion that I've seen a bot buy a masquerade.

Unfortunately the bots only reach a golden deck by accident and usually try to accumulate too many green cards. The best they can do is typically to play out all the actions from hand with the bishop last, trashing whatever else is in hand. This might still let a bot trash a gold and buy a gold, say, which can mimic a good strategy.

[LastFootnote]

Some of the AI for specific cards is just blatantly sub-par. The biggest example I can think of is Salvager. The bot will always prioritize salvaging Copper over Estate, which is just awful.

[brokoli]

I've never seen the bot buying a cultist. It's frustrating because I would like to know what is the effect of ruins in my deck ! ^^

[DG]

The bots do buy cultists but they are not always top of their priority list. Since the bots always try to purchase the most expensive card possible, if you only have one 5 cost card in the kingdom then the bots will probably buy it: cultist, outpost, counting house, whatever.

[aaron0013]

Some of the AI for specific cards is just blatantly sub-par. The biggest example I can think of is Salvager. The bot will always prioritize salvaging Copper over Estate, which is just awful.

It depends on the kingdom.

[jotheonah]

Some of the AI for specific cards is just blatantly sub-par. The biggest example I can think of is Salvager. The bot will always prioritize salvaging Copper over Estate, which is just awful.

It depends on the kingdom.

Baron/Crossroads deck?

[aaron0013]

Some of the AI for specific cards is just blatantly sub-par. The biggest example I can think of is Salvager. The bot will always prioritize salvaging Copper over Estate, which is just awful.

It depends on the kingdom.

Baron/Crossroads deck?

And Silk Roads when you are randomly masqueraded a Salvager from an unsuspecting opponent.

[SirPeebles]

Wouldn't it be better to simply not play the Salvager at all in the Silk Roads example?  Unless maybe you aren't rushing them, but rather planning some sort of megaturn instead.

[jotheonah]

Wouldn't it be better to simply not play the Salvager at all in the Silk Roads example?  Unless maybe you aren't rushing them, but rather planning some sort of megaturn instead.

You just played some Apothecaries or a Counting House and you have 9 coppers in hand. You're trashing one with the Salvager in order to buy 2 Silk Roads.

[aaron0013]

As I said, it depends on the kingdom.

[SirPeebles]

That's more than depending on the kingdom.  It's depending on a particular predicament along a specific oddball strategy.  How frequently does one purchase apothecaries or counting houses while pursuing a silk road strategy, while your opponent is sending you a salvager with a masquerade?  Even with all of those cards in the kingdom...

[jotheonah]

I might snag a Counting House in SR game.

[aaron0013]

Hooray for being overly technical

[Toskk]

I've been playing against the various Goko bots quite a lot, although just with base + Intrigue, so I haven't had the chance to see their behavior with many of the trash cards mentioned.. however I can definitely say that most of the bots play Masquerade very poorly, passing Coppers over Estates in most cases. A few of the higher-rated bots do it correctly, though.

In my opinion, probably the area the bots are weakest on is when to go for Duchies and Estates. They generally start too late. For example, most of them will spot a Workshop/Ironworks + Garden combo, and play it perfectly.. up until it comes time to finish out the Estate pile.. but when they see that they're tied on points, with just a few Estates left, they switch to Silver to try to grab a Duchy or two, and it's too late for that.

Similarly, they tend to continue buying cost-5/6 cards over Duchies sometimes even with as few as four Provinces left (occasionally even fewer than that).

[DG]

I've actually just come across the best play I've seen yet from the bots. It started with masquerade, buying spies then peddlers, adding a ghost ship and then a few king's courts. It seemed to play the masquerade well and was able to get consistent heavy attacks going with province buying hands from king's court on peddlers.

[aaron0013]

That would be an interesting strategy. Kings Court a Peddler or Spy, play Ghost Ship, then Kings Court another Spy in case he put good cards back on top. It would probably force him to always top deck bad cards, which could be profitable most cases. Of course, there are a lot of other things you could be doing with your KCs.....

[timchen]

I've actually just come across the best play I've seen yet from the bots. It started with masquerade, buying spies then peddlers, adding a ghost ship and then a few king's courts. It seemed to play the masquerade well and was able to get consistent heavy attacks going with province buying hands from king's court on peddlers.

Isn't the good strategy here is to pin you to death?

[DG]

I'll be very glad if the bots are not coded with masquerade pins.

[thirtyseven]

(Base only) game where Warlord Bot uses TR/Workshop to help him piledrive Moneylender!

[aaron0013]

Bot:  What's wrong with moneylender? I like it!

[SirPeebles]

With a hand of 5 Moneylenders you can buy a Copper.  This is what we call synergy.

[thirtyseven]

Buying nothing with opening $2 and Chapel and Cellar on the board I thought was interesting:

------------ Game Setup ------------
Supply cards: Cellar, Chapel, Moat, Village, Workshop, Moneylender, Spy, Throne Room, Council Room, Library, Copper, Silver, Gold, Estate, Duchy, Province, Curse
guest8769.0000 - starting cards: Estate, Estate, Estate, Copper, Copper, Copper, Copper, Copper, Copper, Copper
Lord Bottington - starting cards: Estate, Estate, Estate, Copper, Copper, Copper, Copper, Copper, Copper, Copper
guest8769.0000 - shuffles deck
Lord Bottington - shuffles deck
guest8769.0000 - draws Copper, Copper, Copper, Copper, Copper
Lord Bottington - draws Estate, Estate, Estate, Copper, Copper
 
---------- guest8769.0000: turn 1 ----------
guest8769.0000 - plays 5 Copper
guest8769.0000 - gains Silver
guest8769.0000 - draws Copper, Estate, Copper, Estate, Estate
 
---------- Lord Bottington: turn 1 ----------
Lord Bottington - plays 2 Copper
Lord Bottington - draws Copper, Copper, Copper, Copper, Copper

[ftl]

Yeah, Cellar+Library would seem like the obvious right opening to me there, they go well together.

[gordythegopher]

Don`t know whether this counts as bizzare, but the bots just don`t know when they`re beaten. After clearing out 2 piles, and with only 1 worker`s village left, Defender Bot just kept stubbornly drawing victory point cards instead of ending the game. The game finally ended when I`d emptied the coppers pile, which allowed me to amass 219 points. And I`m not even a good player as you can probably tell....

[Drab Emordnilap]

Don`t know whether this counts as bizzare, but the bots just don`t know when they`re beaten. After clearing out 2 piles, and with only 1 worker`s village left, Defender Bot just kept stubbornly drawing victory point cards instead of ending the game. The game finally ended when I`d emptied the coppers pile, which allowed me to amass 219 points. And I`m not even a good player as you can probably tell....

I think it's entirely reasonable to code an AI to never take an action that will result in a guaranteed loss.

[DG]

I think it's entirely reasonable to code an AI to never take an action that will result in a guaranteed loss.

I raised this problem in Beta for the bots not finishing games. You could argue that bots shouldn't ever buy the last card to lose a game, however the game is being run for the benefit of the human customer. If the bot creates a tedious protracted endgame when it has no mathematical chance of winning the game, except by the human paymaster quitting in boredom, then it should instead lose graceful and quickly. There is nothing to be gained for anyone from bot stubbornness. (In some earlier versions the bot would actually buy a card for a close loss when continuing might have found a win, but persevere forever in games that were entirely lost.)

In multiplayer games the situation is more complex but if the bots employ a simple strategy of maximizing their score then they avoid king making problems.

[Drab Emordnilap]

You could argue that bots shouldn't ever buy the last card to lose a game, however the game is being run for the benefit of the human customer. If the bot creates a tedious protracted endgame when it has no mathematical chance of winning the game, except by the human paymaster quitting in boredom, then it should instead lose graceful and quickly.

If it's so easy for the bot to end the game, why is it impossible for the human player to end the game instead?

In multiplayer games the situation is more complex but if the bots employ a simple strategy of maximizing their score then they avoid king making problems.

I think there's a few different ways they can go here -- one can aim for maximum personal score; one can shoot for improving one's placement relative to others, or one can go for first place. If I'm in a 3 player game and I can end the game now by buying the last Province, putting me from 3rd to 2nd place, I'd be much more inclined to try and play for first still. It depends on what the intended goal is.

[DG]

If it's so easy for the bot to end the game, why is it impossible for the human player to end the game instead?

Once you've got 11 provinces in your deck you might not find it easy to buy the twelfth. I come across this problem against the bots more often then you might think.

[Taco Lobster]

Someone mentioned this in another thread, and I tried it last night to great amusement: using Masquerade to give a bot Rats.  The bot has no idea what to do with Rats, and just plays it because it's a cantrip.  Within a few turns, the bot will have a deck of all Rats and the one Copper it can buy each turn.

[theory]

Someone mentioned this in another thread, and I tried it last night to great amusement: using Masquerade to give a bot Rats.  The bot has no idea what to do with Rats, and just plays it because it's a cantrip.  Within a few turns, the bot will have a deck of all Rats and the one Copper it can buy each turn.

The thematic implications of this are hilarious.

[aaron0013]

What? A taco lobster masquerading as a rat?

[Morgrim7]

Many of the bots I've played seem to choose random strategies out of the blue, like rushing Moneylenders.

[jotheonah]

So my coding experience is all of one class in college on javascript, but it seems like the Goko bots are already coded with a brute force code that thinks things like "IF Cantrip or Money, THEN Buy as many as you can."  It doesn't seem like rocket science to have a "Buy 1" category for cards like Moneylender and Chapel.

The big problem I've noticed is they don't know/care about their money density. So when they get into a position where they consistently hit less than 3, they buy Coppers and Estates, making their deck worse and worse. Or they buy lousy 3's that don't help their deck value over Silver.

[Kuildeous]

The big problem I've noticed is they don't know/care about their money density. So when they get into a position where they consistently hit less than 3, they buy Coppers and Estates, making their deck worse and worse. Or they buy lousy 3's that don't help their deck value over Silver.

Although, coding the AIs to base their buying decisions on money density could be pretty nasty, since what is tricky for humans to track would be a doddle for a computer to do.

It wouldn't be as simple as saying "If the new card does not increase my average coin, then don’t buy it," because victory cards do decrease your average coin and will need to be bought at some point.

I suppose there could just be a tipping point. The bot can be programmed to increase average coin and then when it goes green, that conditional no longer applies.

[jotheonah]

That is a good point.  I think there ought to be some money density it won't go below ever, if the idea is to force them to never get into negative feedback loops where all they can afford are Estates so they just buy more and more Estates.

[WallyNate]

In a recent game, the bot ruthlessly chapel'd down his deck until he only had a silver and pearl divers left.  Then every single turn, he played every pearl diver he had, then bought another pearl diver.  When the pearl divers finally ran out, he bought a copper to start building up again.

[Beyond Awesome]

In a recent game, the bot ruthlessly chapel'd down his deck until he only had a silver and pearl divers left.  Then every single turn, he played every pearl diver he had, then bought another pearl diver.  When the pearl divers finally ran out, he bought a copper to start building up again.

I hate bot games like that. They are long and tedious and you are essentially playing solitaire at that point.

[Schneau]

In a recent game, the bot ruthlessly chapel'd down his deck until he only had a silver and pearl divers left.  Then every single turn, he played every pearl diver he had, then bought another pearl diver.  When the pearl divers finally ran out, he bought a copper to start building up again.

Pearl Diver is so OP. I'm sure you lost this game.

[SirPeebles]

Someone mentioned this in another thread, and I tried it last night to great amusement: using Masquerade to give a bot Rats.  The bot has no idea what to do with Rats, and just plays it because it's a cantrip.  Within a few turns, the bot will have a deck of all Rats and the one Copper it can buy each turn.

Here we are, a month and a half later and if Rats end up in a bot's deck it's game over for them.  And apparently if I switch tabs over to here so that I can complain to people who aren't responsible for it, the bot waits for me to tab back before continuing to play the 20 Rats.  It's as though Goko wants to discourage from complaining.

[Lashof]

In a recent game, the bot ruthlessly chapel'd down his deck until he only had a silver and pearl divers left.  Then every single turn, he played every pearl diver he had, then bought another pearl diver.  When the pearl divers finally ran out, he bought a copper to start building up again.

This also happens with Chapel + Spy.

Regarding it waiting while you switch tabs:  If you actually use a separate window instead of a separate tab, it continues to run.

[LastFootnote]

My favorite is when it continues to buy Warehouses whenever it hits $3, always plays Warehouses, and never discards Warehouses when playing a Warehouse. The bot just plays 4 or 5 Warehouses in a turn and buys nothing. Classic!

[Kirian]

The big problem I've noticed is they don't know/care about their money density. So when they get into a position where they consistently hit less than 3, they buy Coppers and Estates, making their deck worse and worse. Or they buy lousy 3's that don't help their deck value over Silver.

Although, coding the AIs to base their buying decisions on money density could be pretty nasty, since what is tricky for humans to track would be a doddle for a computer to do.

It wouldn't be as simple as saying "If the new card does not increase my average coin, then don’t buy it," because victory cards do decrease your average coin and will need to be bought at some point.

I suppose there could just be a tipping point. The bot can be programmed to increase average coin and then when it goes green, that conditional no longer applies.

Actually, consider the code written for the simulators out there.  Not only is this relatively easy to do for a bot, it's readily codeable.

[Donald X.]

My favorite is when it continues to buy Warehouses whenever it hits $3, always plays Warehouses, and never discards Warehouses when playing a Warehouse. The bot just plays 4 or 5 Warehouses in a turn and buys nothing. Classic!

I have reported this one a couple times.

I am told that they have an improved AI in-house; dunno if it fixes this.

[pinkymadigan]

My favorite is when it continues to buy Warehouses whenever it hits $3, always plays Warehouses, and never discards Warehouses when playing a Warehouse. The bot just plays 4 or 5 Warehouses in a turn and buys nothing. Classic!

I have reported this one a couple times.

I am told that they have an improved AI in-house; dunno if it fixes this.

Good thing they are keeping it to themselves. We wouldn't want that going public.

[jotheonah]

The big problem I've noticed is they don't know/care about their money density. So when they get into a position where they consistently hit less than 3, they buy Coppers and Estates, making their deck worse and worse. Or they buy lousy 3's that don't help their deck value over Silver.

Although, coding the AIs to base their buying decisions on money density could be pretty nasty, since what is tricky for humans to track would be a doddle for a computer to do.

It wouldn't be as simple as saying "If the new card does not increase my average coin, then don’t buy it," because victory cards do decrease your average coin and will need to be bought at some point.

I suppose there could just be a tipping point. The bot can be programmed to increase average coin and then when it goes green, that conditional no longer applies.

Actually, consider the code written for the simulators out there.  Not only is this relatively easy to do for a bot, it's readily codeable.

I find coding problems highly interesting on a very abstract level, but my actual knowledge of code is negligible.

[DG]

Of all the dodgy bot strategies I've seen, I feel that using a rebuild to trash a duchy and gain an estate is probably the worst .

[ftl]

Hah. I can even imagine why that happens. The Bot correctly figures that if it gains the last province then it loses, and so it gains an estate instead since Duchies are out?

[DG]

No. With half a dozen estates, duchies, and provinces in the supply it decided to rebuild a duchy into an estate, twice. It had been successfully rebuilding some estates into estates prior to that, so this was obviously an ambitious leap forward.

[clb]

No. With half a dozen estates, duchies, and provinces in the supply it decided to rebuild a duchy into an estate, twice. It had been successfully rebuilding some estates into estates prior to that, so this was obviously an ambitious leap forward.

Clearly, first place is better than third, which is better than 6th, so why wouldn't you want more cards with a big 1 on them?

[jotheonah]

Not satisfied being merely ineffective, Goko bots start actively sabotaging their own scores!

[werothegreat]

Whatever happened to Lord Bottington's diary?

[SirPeebles]

Heh, a bot just Pillaged my Scout.

[Kuildeous]

Heh, a bot just Pillaged my Scout.

Didn't even buy him dinner first? That's rude.

[SirPeebles]

The Scout just kept looking for someone named Josephine.

[Avalanchian]

Someone mentioned this in another thread, and I tried it last night to great amusement: using Masquerade to give a bot Rats.  The bot has no idea what to do with Rats, and just plays it because it's a cantrip.  Within a few turns, the bot will have a deck of all Rats and the one Copper it can buy each turn.

I just mentioned this in stupid pet tricks before realising this thread existed. Turns out this trick is already well established.

I agree - this is hilarious.

[Joseph2302]

Someone mentioned this in another thread, and I tried it last night to great amusement: using Masquerade to give a bot Rats.  The bot has no idea what to do with Rats, and just plays it because it's a cantrip.  Within a few turns, the bot will have a deck of all Rats and the one Copper it can buy each turn.

Yes this is a definite issue with the bots, also seen them frequently with 3-4 Golems but only 2-3 other actions in deck!!!

[Avalanchian]

I was playing a game against Lord Bottington today and had a really meh kingdom using Colonies (Pawn, Great Hall, Masquerade, Shanty Town, Mining Village, Quarry, Thief, Counting House, Upgrade, King's Court). I can't see anything too great here, but I'd love to hear from anyone who sees something I've probably missed. I basically went Masq + BM, which is the kind of style I tend to shy away from in province games, let alone Colonies. Lord Bottington certainly couldn't see anything here, but rather than going BM (which I'm assuming they've been programmed to try and avoid to add variety) he spent the ENTIRE game buying low cost actions and upgrading them one by one. As you can imagine, the game ended 80-19.

I thought this turn summed up his dutifully aimless style nicely (starting with KC, Upgrade, Upgrade, Silver, Copper):

---------- Lord Bottington: turn 17 ----------
Lord Bottington - plays King's Court
Lord Bottington - plays Upgrade
Lord Bottington - draws Mining Village
Lord Bottington - trashes Mining Village
Lord Bottington - gains Duchy
Lord Bottington - plays Upgrade
Lord Bottington - draws Upgrade
Lord Bottington - trashes Silver
Lord Bottington - gains Mining Village
Lord Bottington - plays Upgrade
Lord Bottington - shuffles deck
Lord Bottington - draws Upgrade
Lord Bottington - trashes Upgrade
Lord Bottington - gains Gold
Lord Bottington - plays Upgrade
Lord Bottington - draws Silver
Lord Bottington - trashes Upgrade
Lord Bottington - gains Gold
Lord Bottington - plays 1 Copper, 1 Silver
Lord Bottington - buys Silver
Lord Bottington - gains Silver
Lord Bottington - draws Pawn, Silver, Masquerade, Mining Village, Thief

[ftl]

There's cute stuff to be had with Counting House/Pawn. I don't know how good it actually is, but for a colony game I'd try it. Dunno if it beats masq-BM for colonies.

[SirPeebles]

I would probably open Masq-Quarry and go for an Upgrade and King's Court strategy.  Upgrade -> Gold -> King's Court -> Province looks strong, particularly since you can play King's Court on the Upgrade.

[Avalanchian]

There's cute stuff to be had with Counting House/Pawn. I don't know how good it actually is, but for a colony game I'd try it. Dunno if it beats masq-BM for colonies.

Even after saying that, it took me a good while to work out why that particular combo might work. Use the pawns for non-terminal +buy to grab a ton of copper before using Counting house to bring it all back to your hand, right?

I would probably open Masq-Quarry and go for an Upgrade and King's Court strategy.  Upgrade -> Gold -> King's Court -> Province looks strong, particularly since you can play King's Court on the Upgrade.

I considered this at the time, but it all felt too slow and wasn't sure how useful upgrading to Provinces was in a Colony game (I'm still pretty new to the feel of Colony games).

Thank you both for the tips. I really need to get more creative when I see a kingdom and just accept that I'm going to lose a lot as I get a feel for this creativity >_<

[SirPeebles]

Oh, I hadn't noticed it was a Colony game.  It might still work?  You are unlikely to grab any Colonies though.

In general, in a Colony game you want to look for an engine.  Moreover with King's Court available, it seems even more likely that there should be an engine.  Masq is a great opening trasher, and King's Court fills the $7 gap that often stops Upgrade from gaining multiple Provinces per turn.

One possibility, if you get a sleek KC/Upgrade engine going, is to Upgrade past Province for Platinum and then buy Colonies.  Ordinarily I would think Upgrade is too slow for this, but with King's Court maybe not?

[SirPeebles]

Well, my strategy beat Lord Bottington, but then Lord Bottington didn't do much of anything.  He got 2 Provinces and 0 Colonies in 17 turns.

[Avalanchian]

Okay, I don't want to end up hijacking this thread with this kingdom - but this is insightful for me dammit! >_<

Well, my strategy beat Lord Bottington, but then Lord Bottington didn't do much of anything.  He got 2 Provinces and 0 Colonies in 17 turns.

Yeah that's what he had in my game after 21 turns. Looks like your KC+Upgrade strategy is faster (that or I played my BM badly, though I did feel I had excellent shuffle luck). I'll try it myself now to see if I have the technique to pull it off. Thanks for your help =)

[SirPeebles]

Okay, I don't want to end up hijacking this thread with this kingdom - but this is insightful for me dammit! >_<

Well, my strategy beat Lord Bottington, but then Lord Bottington didn't do much of anything.  He got 2 Provinces and 0 Colonies in 17 turns.

Yeah that's what he had in my game after 21 turns. Looks like your KC+Upgrade strategy is faster (that or I played my BM badly, though I did feel I had excellent shuffle luck). I'll try it myself now to see if I have the technique to pull it off. Thanks for your help =)

I tried a pure BM Masq strategy last night, and I got to 6 Colonies and 1 Province in 19 turns I believe.  I was slowed down a bit by some KCed Thieves who stole lots of Plats and Golds.

Also, I made this topic, and I'm all for whatever discussion comes from it.

[Avalanchian]

I tried a pure BM Masq strategy last night, and I got to 6 Colonies and 1 Province in 19 turns I believe.  I was slowed down a bit by some KCed Thieves who stole lots of Plats and Golds.

Also, I made this topic, and I'm all for whatever discussion comes from it.

The BM+Masq is weirdly fast here. I can't seem to hone the engine down as quickly as you either. I guess that's what experience does huh? I've tried several approaches, but all seem to take me longer than my first game which was simply Masq+BM (I don't think it helps that the bot seems to go Thief on me now when I start getting Platinum, which seems to slow me down =/). Clearly I'm missing a trick (or four).

Well I'm glad you're all for this conversation. It's kind of turned into "let's teach Ava to build an engine" though =P

[SirPeebles]

Well, for the big money I bought two Masquerades.   Are you getting two?  Trash Estates freely, but count your money before trashing Copper; if you're at $9 exactly, you're better off buying Platinum than trashing Copper.  I think in the game I played, I hit $11 pretty early, but opted to pick up my first Platinum rather than snag the early Colony, since that Platinum will help me get more Platinum more quickly.

Big Money is often worse in Colony games since all that starting Copper is so much less valuable than in a Province game, but Masq is one of the few strong BM enablers that things out your Copper.

[SirPeebles]

Speaking of building engines, if you haven't already read Geronimoo's Article I strongly recommend it.

[Avalanchian]

Yeah, I'm actually getting worse. That or I'm over-thinking this or I'm getting board fatigue now... I keep losing to KC+Thief trashing the Platinum out of my thin deck. Pretty embarrassing...

Speaking of building engines, if you haven't already read Geronimoo's Article I strongly recommend it.

Yeah, I spent pretty much a whole weekend with that article and that kingdom trying to make the engine work, playing against myself doing Smithy+BM to measure my technique. I managed to get it working reasonably consistently it in the end, but had to cut back on the recommended markets. I found spending that many $5's on markets really slowed getting my other engine pieces. I learned a lot but enough it seems >_>

Maybe I should give it another look....

[Avalanchian]

With no +buy, no trashing, no attack and Hoard/Bank/Embassy on the board, BM looked good. Lord Bottington scoffs at BM though, and constructed this hilarious work of art.

Turns out the bots also enjoy Scout jokes.

[DG]

I've played a game where despite no silver or gold being bought, two warlord bots bought all the noble brigands. Fortunately I had a chapel and was able to clean out my deck and continue with a highways, buying out a few provinces before the coppers ran out.

[Joseph2302]

Getting the feeling some bots like to overbuy cards: played a game where they bought 9 apothecaries (in a deck of c.20 cards), and then this one (http://dominionlogs.goko.com/20130214/log.505f2ab2f93b9113b758d65c.1360873560878.txt), where they apprenticed their deck to 5 cards and just kept it there!!! silly bots

[GendoIkari]

Heh, a bot just Pillaged my Scout.

This is a thread for bad bot strategies, not bad human strategies. Post about games where you bought a Scout elsewhere!

[heron]

Avalanchian: Since it's a bot and I don't have to worry about it being too good, I might try KC-KC-Thief-Thief-Masq just for fun.

[DG]

---------- Lord Bottington: turn 3 ----------
 Lord Bottington - plays Feast
 Lord Bottington - trashes Feast
 Lord Bottington - gains Feast
 Lord Bottington - plays 3 Copper
 Lord Bottington - buys Silver
 Lord Bottington - gains Silver
 Lord Bottington - draws Copper, Copper, Estate, Copper, Loan

[Avalanchian]

Avalanchian: Since it's a bot and I don't have to worry about it being too good, I might try KC-KC-Thief-Thief-Masq just for fun.

Oohh that does seem fun. I might give that one a try some time.

---------- Lord Bottington: turn 3 ----------
 Lord Bottington - plays Feast
 Lord Bottington - trashes Feast
 Lord Bottington - gains Feast

Clearly trying to 3-pile for the win without clogging the deck with endless terminals. Very clever

[hsiale]

I just had Warlord Bot end the game with 10 Potions in his deck but not buying a Potion cost card (Possession was the only one available). It was a decent defense against my Possession (I once played it with KC and ended up getting nothing due to terrible hands I was given) but nothing more.

[DG]

The bots are awful with potions at the moment and this is probably why alchemy is still in beta.

I did have another one today don't have the log. Very promising play.

Turn 3, bot plays young witch, spends 3 coppers, buys a watchtower, reveals a watchtower, trashes the watchtower.

[spiritbears]

long time lurker
Yes^  I notice they have such a penchant for opening with potion, even if the only alch card is vinyard!  If there is any potion requiring card at all, they open potion on 4/3. 
But, still, goko has come a long way and I am sure happy to have the entire array of cards up and working (for the most part)!  I was extremely skeptical, but couldn't be happier!
Can't wait for the iOs!
-Spirit Bears

[Avalanchian]

Banker Bot   plays Margrave
Banker Bot   draws Gold, Crossroads, Estate
Avalanchian    draws Province
Avalanchian    discards Province
Avalanchian    discards Copper
Avalanchian    discards Margrave
Banker Bot   plays 3 Gold
Banker Bot   buys Province
Banker Bot   buys Survivors
Banker Bot   draws Silver, Silver, Farming Village, Gold, Gold

Gotta use dem +buys...

[Avalanchian]

http://dom.retrobox.eu/?/20130304/log.50e87c23e4b0101dec0f5c9d.1362436947165.txt

Conqueror Bot   plays Jester
Avalanchian   reveals Curse
Avalanchian   discards Curse
Conqueror Bot   gains Curse

Optimal Jester play right there. I can't think of any reason why the bot may have done that. More variety in the deck to be more likely to activate menagerie?

The bot's final deck was: 2 Menagerie, 2 Jester, 1 Great Hall, 1 Minion, 9 Copper, 5 Gold, 1 Silver, 5 Estate, 2 Province, 1 Duchy, 1 Curse

[Squidd]

On the one hand, it's always fun to make a Hunting Party deck that can easily absorb all 8 Provinces. On the other hand, it's less satisfying when the reason you take all 8 is that the bot:

- Opens HP, Chapel
- Trashes all starting cards while buying Silver, Silver, Potion
- Buys 10 Scrying Pools, playing every SP in the deck every turn
- With one Province remaining, buys an Estate

[spiritbears]

On the one hand, it's always fun to make a Hunting Party deck that can easily absorb all 8 Provinces. On the other hand, it's less satisfying when the reason you take all 8 is that the bot:

- Opens HP, Chapel
- Trashes all starting cards while buying Silver, Silver, Potion
- Buys 10 Scrying Pools, playing every SP in the deck every turn
- With one Province remaining, buys an Estate

Yes!  I've seen more than one bot do that^!

[werothegreat]

http://dom.retrobox.eu/?/20130304/log.50e87c23e4b0101dec0f5c9d.1362436947165.txt

Conqueror Bot   plays Jester
Avalanchian   reveals Curse
Avalanchian   discards Curse
Conqueror Bot   gains Curse

Optimal Jester play right there. I can't think of any reason why the bot may have done that. More variety in the deck to be more likely to activate menagerie?

The bot's final deck was: 2 Menagerie, 2 Jester, 1 Great Hall, 1 Minion, 9 Copper, 5 Gold, 1 Silver, 5 Estate, 2 Province, 1 Duchy, 1 Curse

I've seen bots do that several times.

[SirPeebles]

Mint and Fool's Gold were both on the board, and the bot opened with Mint.  I was rather impressed that it knew to go for Mint/FG.  But then T2 it bought a Chapel 

[A Drowned Kernel]

During a base set adventure I saw a bot chapel its deck down until its only treasures were a silver and a gold. Then it spent ten turns piledriving the laboratories, one by one.

Then it bought a remodel.

Then it remodeled its gold into a Province.

Then it remodeled its silver into a Duchy.

By the time I finally managed to empty the Province pile (I hadn't built my deck assuming I would have to buy almost all of them, silly me) it was buying a copper and remodeling a copper into an estate every turn.

The thing is that the adventures are supposed to teach new players about the game, so I don't know if this sort of thing is a bug or a feature.

[Powerman]

Yeah, and I've found them to piledrive the spies very often in adventure mode...

[Master Shuffler]

In the single most annoying game ever played on Goko, I present

http://dominionlogs.goko.com//20130403/log.5143c778e4b0d173cd4c231a.1365038040968.txt

I went for big money / chapel and eventually crossroads, but the bot kept insisting scrying pool was good so it bought 10 of them, as well as 10 spies and pretty much nothing else save for 2 fairgrounds. Aside from choosing a horrible strategy, it made the game take way longer than it should have.

[Axxle]

In the single most annoying game ever played on Goko, I present

http://dominionlogs.goko.com//20130403/log.5143c778e4b0d173cd4c231a.1365038040968.txt

I went for big money / chapel and eventually crossroads, but the bot kept insisting scrying pool was good so it bought 10 of them, as well as 10 spies and pretty much nothing else save for 2 fairgrounds. Aside from choosing a horrible strategy, it made the game take way longer than it should have.

If that happens you can probably turn the animation speed all the way up

[Kirian]

In the single most annoying game ever played on Goko, I present

http://dominionlogs.goko.com//20130403/log.5143c778e4b0d173cd4c231a.1365038040968.txt

I went for big money / chapel and eventually crossroads, but the bot kept insisting scrying pool was good so it bought 10 of them, as well as 10 spies and pretty much nothing else save for 2 fairgrounds. Aside from choosing a horrible strategy, it made the game take way longer than it should have.

If that happens you can probably turn the animation speed all the way up

People play with the animation speed at any other setting?

[loppo]

I used to play with speed settings at fastest. There was an update in January where the animations were adjusted and since then i switched back to normal. Now i have no troubles following the plays on other persons turns, and game speed feels ok for me.

But i'm not a speed freak at all. I enjoy the game, and i can live with playing a game less per day.

[Squidd]

I had a bot that kept playing all its Schemes, buying a Scheme, and then scheming the Schemes, over and over, until it had them all. Quicker to resolve than SP/Spy, but arguably dumber, because it was literally the same as having none of them.

[DStu]

Quicker to resolve than SP/Spy, but arguably dumber, because it was literally the same as having none of them.

In before edge case

[Morgrim7]

I had a bot that kept playing all its Schemes, buying a Scheme, and then scheming the Schemes, over and over, until it had them all. Quicker to resolve than SP/Spy, but arguably dumber, because it was literally the same as having none of them.

Edge case: You are behind and your deck is bloated with coppers, curses and ruins from various junkers. Besides crap, your deck has exactly 5 highways and 5 GMs. In order to pair them up for a megaturn, you use your schemes to find them all one by one and return them all to the top of your deck.

[Squidd]

There was an implicit "in context" there. In context, Scheme was doing nothing.

[Squidd]

Also, for every mega-turn piece you find, that's one less Scheme you can topdeck per turn. You're going to run out of Schemes way before you pull the mega-turn together.

[eHalcyon]

Also, for every mega-turn piece you find, that's one less Scheme you can topdeck per turn. You're going to run out of Schemes way before you pull the mega-turn together.

KC-Scheme lets you topdeck KC, Scheme, and a third thing.

[eHalcyon]

Quicker to resolve than SP/Spy, but arguably dumber, because it was literally the same as having none of them.

In before edge case

Actual edge case for why repeatedly playing and topdecking Schemes is different from not having them at all (note that using Scheme to topdeck another card is not an edge case, because that is its actual purpose):

- use them to lower the price of Peddler
- use them to activate Conspirators
- bump up the value of HoP
- Scheme is the bane for YW
- a Scheme in hand takes up space and helps you activate Menagerie (e.g. if not having that Scheme would have meant having a duplicate in hand)
- fodder for TfBs

Some of these examples can only explain the topdecking of a single Scheme -- for example, it would be ridiculous to topdeck more than 1 Scheme when it is the bane.  One is enough!  But never fear, there is an edge case to this edge case -- Minion attacks.  If you only top deck a single Scheme, Minion will remove your protection!  But if you topdeck six Schemes, you are protected for at least that turn.

[DG]

The warlord bot horribly overbuys schemes. I play with just base and hinterlands so I see it quite often. Once two warlord bot opponents start buying schemes and a noble brigand it becomes really tempting to abandon the game.

[WanderingWinder]

In the single most annoying game ever played on Goko, I present

http://dominionlogs.goko.com//20130403/log.5143c778e4b0d173cd4c231a.1365038040968.txt

I went for big money / chapel and eventually crossroads, but the bot kept insisting scrying pool was good so it bought 10 of them, as well as 10 spies and pretty much nothing else save for 2 fairgrounds. Aside from choosing a horrible strategy, it made the game take way longer than it should have.

If that happens you can probably turn the animation speed all the way up

People play with the animation speed at any other setting?

I'm pretty well known for playing quickly, no? I play with it on fast (or whatever they're calling the third setting, second from the fastest, now).

[elahrairah13]

I played an Androminion game this week in which the bot bought several loans and therefore trashed down to just a handful of coppers...
but also bought 4 or 5 Counting Houses.

[ConMan]

When I play Androminion and Rats is on the board, well - their deck is certainly safe against Curses.

[eHalcyon]

When I play Androminion and Rats is on the board, well - their deck is certainly safe against Curses.

If Androminion bot picks up Rats, I quit.

[ConMan]

When I play Androminion and Rats is on the board, well - their deck is certainly safe against Curses.

If Androminion bot picks up Rats, I quit.

So would I, but I recently installed it on a new phone and I need the "10 wins in a row" achievement again.

For those who don't know how painful it can be, Androminion does a little cascading display of cards gained or revealed, so when your opponent's deck is 20 copies of Rats it's a huge waterfall as each play of Rats results in a reveal of up to 5 more copies.

[Blueswan]

I won the 20th stage of Cornucopia adventures with a "Moat" strategy. I ended with 8VP, the three bots were all in zero or negative territory.

http://dominionlogs.goko.com//20130405/log.50f70ca7e4b0c53c94b0ca2d.1365155526139.txt

[LastFootnote]

I won the 20th stage of Cornucopia adventures with a "Moat" strategy. I ended with 8VP, the three bots were all in zero or negative territory.

http://dominionlogs.goko.com//20130405/log.50f70ca7e4b0c53c94b0ca2d.1365155526139.txt

I love that about the Adventure finals. If I recall correctly, the final stage of Intrigue Act 3 is a board where Secret Chamber is the key card. It's pretty sweet.

EDIT: Assuming I'm remembering the correct stage, it's a four-player game with this board:

$2: Chapel, Moat, Pawn, Secret Chamber
$3: Shanty Town, Swindler, Village
$5: Torturer, Tribute, Witch

[JohnO]

All the bots had a major AI upgrade this week, one is rated over 4k now, they were about 3400  before I think.

[hsiale]

Bots definitely don't understand what is the bane card and how it influences their decision if they should buy Young Witch.

I played two YW games today. In the first one there was Squire (as bane), Lighthouse and Vineyard (and Market+Festival to strengthen the Vineyard strategy even more). First thing I did was piledriving Squires, with Lighthouses after that. Still the bot went heavy YW (3 or 4 of them). I ended the game with 8 Vineyards worth 14 points each and all 10 Curses still in the Supply.

The second game had Oasis as bane and Watchtower in addition to this, which not only was a good YW defense, but also protected from Ghost Ship and allowed a "trash Copper, trash Copper, topdeck Cache" trick. This time at least 4 YWs managed to give me 4 Curses, half of which I trashed with Apprentice.

[clb]

Bots definitely don't understand what is the bane card and how it influences their decision if they should buy Young Witch.

I played two YW games today. In the first one there was Squire (as bane), Lighthouse and Vineyard (and Market+Festival to strengthen the Vineyard strategy even more). First thing I did was piledriving Squires, with Lighthouses after that. Still the bot went heavy YW (3 or 4 of them). I ended the game with 8 Vineyards worth 14 points each and all 10 Curses still in the Supply.

The second game had Oasis as bane and Watchtower in addition to this, which not only was a good YW defense, but also protected from Ghost Ship and allowed a "trash Copper, trash Copper, topdeck Cache" trick. This time at least 4 YWs managed to give me 4 Curses, half of which I trashed with Apprentice.

It has been a little while, so perhaps has been fixed, but the last YW game I played against a bot featured Scheme as the bane. I felt that 1 or 2 were enough (I don't remember the board - I ended up doing some money strategy, I think), but the bot felt it needed all 8 or 9 of the other Schemes.

[hsiale]

This might have been not due to Scheme being the bane, but due to it being a cantrip. Goko bots love cantrips. They piledrive Spy nearly every game it's in the kingdom.

[Blueswan]

This might have been not due to Scheme being the bane, but due to it being a cantrip. Goko bots love cantrips. They piledrive Spy nearly every game it's in the kingdom.

Indeed. And when you lose to a bot who piledrives Spy, that's when you know you played a shitty game. Been there done that.

[Morgrim7]

With Sheme as the bane of YW all you need is one and just topdeck it every turn.

[Taco Lobster]

The bot definitely loves to go village idiot on Scheme.  I don't know how many endgames when Scheme is present come down to the bot spamming Schemes, buying another Scheme, and then returning the played Schemes to the top of its deck. 

[Squidd]

Mint and Fool's Gold were both on the board, and the bot opened with Mint.  I was rather impressed that it knew to go for Mint/FG.  But then T2 it bought a Chapel 

I think the bots (or at least some of them) always take Mint on a 5/2, without regard to the $2 cards. I've seen it a few times. Just now the only $2 was Moat, and the other $5's were Mountebank and Witch.

[clb]

With Sheme as the bane of YW all you need is one and just topdeck it every turn.

True, unless there is some other action I would like to topdeck as it works its way through the deck.

[DG]

Thrones and traders is a great bot strategy.

[hsiale]

http://dom.retrobox.eu/?/20130419/log.50893139a2e67cff211cd7e5.1366394173163.txt - for no visible reason (all Provinces in Supply, no Duke or Silk Road) Lord Bottington buys 3 Duchies in turns 8, 9 an 10.

[D Bo]

I wish I saved the log but in an IGG game this morning one of the bots bought all Warehouses except for two that I smuggled before purchasing any treasure allowing me to pile-drive IGGs and Duchies...

[hsiale]

Bots love Warehouse nearly as much as they love Spy.

[clb]

Bots love Warehouse nearly as much as they love Spy.

+3 cards, +1 action - what's not to love? It's basically a superLab or Village Smithy. How could there be a downside to buying them all?

[Tables]

http://dominionlogs.goko.com//20130424/log.505c7f4ca2e6c78ad2ed5ba1.1366857587510.txt

Village Idiot bot played pretty well. Well, except the part where by endgame, he had 10 potions and no potion cost cards.

[shMerker]

I've noticed bots don't always reveal a Moat to block a Curse when they can. For instance if you play a Witch with a Throne Room they'll block the first curse and not the second one. Is this something they all do or is it limited to lower level bots to make them easier to beat?

[ftl]

I think I remember that being reported as a bug, but it must have been forgotten about and never fixed.

[DG]

Its seems as if everything reported in the invitational beta was lost when they switched systems so they need it reporting again on Get Satisfaction.

[Twistedarcher]

From the last adventure in the base set:


Overlord Hogan IV - plays Remodel
Overlord Hogan IV - trashes Province
Overlord Hogan IV - gains Duchy


Remodeling a province into a duchy! By far the silliest thing I've seen a bot do...

[WanderingWinder]

I've definitely done something like that - however, it was only to immediately end the game (in victory!)

[Twistedarcher]

That would make sense -- but no, there was really no reason to do this. There were still 2 provinces and around 10 duchies left.

[Squidd]

Play Market Square, Market Square, Market Square. Play Count, gaining Copper, trashing hand (full of Copper).

Buy 4 Coppers.

Repeat.

[Squidd]

Not really a "bizarre strategy," but amusing play nonetheless:

I was Bishop-trashing Fortress 7 times a turn most turns (with Menagerie helping the draws). There was at least one turn in which the bot took advantage of my Bishops to also trash his own Fortress 7 times.

[GendoIkari]

From the last adventure in the base set:


Overlord Hogan IV - plays Remodel
Overlord Hogan IV - trashes Province
Overlord Hogan IV - gains Duchy


Remodeling a province into a duchy! By far the silliest thing I've seen a bot do...

Not silly if he had 12 Dukes and 12 Fairgrounds and the right set of other cards... in that case he earns 33 points for that move! (Yes that requires him to have had 12 Dukes and 0 Duchies...)

[D Bo]

From the last adventure in the base set:


Overlord Hogan IV - plays Remodel
Overlord Hogan IV - trashes Province
Overlord Hogan IV - gains Duchy


Remodeling a province into a duchy! By far the silliest thing I've seen a bot do...

Not silly if he had 12 Dukes and 12 Fairgrounds and the right set of other cards... in that case he earns 33 points for that move! (Yes that requires him to have had 12 Dukes and 0 Duchies...)

That would be really impressive if he did that in the Base Adventure set!

[Blueswan]

Just had a fun game against Lord Bottington. It was a kingdom with Swindler, Golem and Rats. I'm sure you can all figure out what happened. The bot bought Golems, I swindled his potion into a Rats and the fun began. Lord Bottington ended the game with 15 Rats and barely any Victory points.

[Blueswan]

Arguably the worst move any bot has played against me:

Conqueror Bot: turn 14
Conqueror Bot   plays Necropolis
Conqueror Bot   plays Minion
Conqueror Bot   discards: Copper, Copper
Conqueror Bot   draws Copper, Province, Copper, Minion
BlueSwan   discards: Copper, Silver, Minion, Overgrown Estate, Province
BlueSwan   draws Copper, Province, Copper, Hovel
Conqueror Bot   plays Minion
Conqueror Bot   discards: Copper, Province, Copper
Conqueror Bot   draws Ambassador, Province, King's Court, Copper
Conqueror Bot   plays King's Court
Conqueror Bot   plays Ambassador
Conqueror Bot   reveals Copper
Conqueror Bot   returns Copper to the Supply
BlueSwan   gains Copper
Conqueror Bot   plays Ambassador
Conqueror Bot   reveals Province
BlueSwan   gains Province
Conqueror Bot   plays Ambassador
Conqueror Bot   reveals Province
BlueSwan   gains Province
Conqueror Bot   draws Gold, Gold, Envoy, Silver, Estate

The bot still had a decent chance of winning before that move: http://dom.retrobox.eu/?/20130522/log.50f70ca7e4b0c53c94b0ca2d.1369209676193.txt

[DStu]

Of course looks as stupid as it could get, but in this case it's not that easy to get the bot to play correctly.
The hand was Prov, Copper, KC, Ambassador, and you have to put in lots of work to play KC correctly with each card.

[ftl]

Hah. Yeah, that Copper needed to not be returned until the third Amb of the KC.

[florrat]

Lol, in the following game a bot bought the last curse in the supply to end the game, which made me win.

http://dom.retrobox.eu/?/20130525/log.510876c7e4b0ac7a7a0bb212.1369516355537.txt

[yed]

Lol, in the following game a bot bought the last curse in the supply to end the game, which made me win.
http://dom.retrobox.eu/?/20130525/log.510876c7e4b0ac7a7a0bb212.1369516355537.txt

It looks like bot was aiming for tie but didn't count turns.

[SCSN]

Lol, in the following game a bot bought the last curse in the supply to end the game, which made me win.

http://dom.retrobox.eu/?/20130525/log.510876c7e4b0ac7a7a0bb212.1369516355537.txt

He clearly didn't have #vpon.

[DG]

http://dominionlogs.goko.com//20130603/log.5083d05fa2e6ac658be2b342.1370319767244.txt

Everything the bots do here is amazingly bad.

---------- Villager Bot: turn 24 ----------
Villager Bot - plays Throne Room
Villager Bot - plays Throne Room
Villager Bot - plays Scheme
Villager Bot - draws Throne Room
Villager Bot - plays Scheme
Villager Bot - draws Throne Room
Villager Bot - plays Throne Room
Villager Bot - plays Throne Room
Villager Bot - plays Throne Room
Villager Bot - plays Scheme
Villager Bot - draws Throne Room
Villager Bot - plays Scheme
Villager Bot - draws Throne Room
Villager Bot - plays Throne Room
Villager Bot - plays Throne Room
Villager Bot - plays Throne Room
Villager Bot - plays Scheme
Villager Bot - draws Estate
Villager Bot - plays Scheme
Villager Bot - draws Silver
Villager Bot - plays Throne Room
Villager Bot - plays Throne Room
Villager Bot - plays Throne Room
Villager Bot - plays 1 Silver
Villager Bot - buys Estate
Villager Bot - gains Estate
Villager Bot - Scheme choice: Throne Room
Villager Bot - Scheme choice: Throne Room
Villager Bot - Scheme choice: Throne Room
Villager Bot - Scheme choice: Throne Room
Villager Bot - Scheme choice: Throne Room
Villager Bot - Scheme choice: Throne Room
Villager Bot - draws Throne Room, Throne Room, Throne Room, Throne Room, Throne Room

[Drab Emordnilap]

Well, that's Villager Bot. There weren't any actual villages, so he had to make do with Throne Rooms.

[Titandrake]

Well, that's Villager Bot. There weren't any actual villages, so he had to make do with Throne Rooms.

It's Marin Bot v0.1, they're still working out the kinks.

[DG]

Sometimes you can't blame the bots for a lack of effort.

---------- Villager Bot: turn 8 ----------
 Villager Bot - plays Village
 Villager Bot - draws Estate
 Villager Bot - plays Village
 Villager Bot - draws Witch
 Villager Bot - plays Witch
 Villager Bot - draws Copper, Witch
 DG - reveals reaction Trader
 DG - gains Silver
 Lord Bottington - reveals reaction Trader
 Lord Bottington - gains Silver
 Villager Bot - plays Witch
 Villager Bot - draws Copper, Curse
 DG - reveals reaction Trader
 DG - gains Silver
 Lord Bottington - reveals reaction Trader
 Lord Bottington - gains Silver
 Villager Bot - plays Witch
 Villager Bot - draws Estate
 Villager Bot - shuffles deck
 Villager Bot - draws Copper
 DG - reveals reaction Trader
 DG - gains Silver
 Lord Bottington - reveals reaction Trader
 Lord Bottington - gains Silver
 Villager Bot - plays 5 Copper
 Villager Bot - buys Witch
 Villager Bot - gains Witch

[DG]

This one is a cracker.

---------- Villager Bot II: turn 33 ----------
 Villager Bot II - duration Tactician
 Villager Bot II - draws Copper, Estate, Estate, Tactician, Copper
 Villager Bot II - plays Border Village Villager Bot II - draws Silver
 Villager Bot II - plays Wandering Minstrel
 Villager Bot II - draws Silver
 Villager Bot II - reveals: Copper, Tactician, Watchtower
 Villager Bot II - places Watchtower on top of deck
 Villager Bot II - places Tactician on top of deck
 Villager Bot II - discards Copper
 Villager Bot II - plays Golem
 Villager Bot II - reveals Tactician, Watchtower
 Villager Bot II - plays Watchtower
 Villager Bot II - plays Tactician
 Villager Bot II - discards: Duchy, Estate, Copper, Estate, Estate, Tactician, Copper, Silver, Silver
 Villager Bot II - draws Estate, Estate
 Villager Bot II - shuffles deck Villager Bot II - draws Estate, Duchy, Copper

[Axxle]

Playing a tactician every turn! Not bad Villager Bot, not bad.

[Tombolo]

Conqueror Bot   cards: 5 Golem, 3 Ironmonger, 14 Potion, 1 Oracle, 1 Mint, 6 Silver, 3 Copper, 1 Masterpiece, 6 Estate, 4 Duchy

He really needed all 14 of those potions to get the crucial 5-5 Golem split!

[Tables]

This one is a cracker.

---------- Villager Bot II: turn 33 ----------
 Villager Bot II - duration Tactician
 Villager Bot II - draws Copper, Estate, Estate, Tactician, Copper
 Villager Bot II - plays Border Village Villager Bot II - draws Silver
 Villager Bot II - plays Wandering Minstrel
 Villager Bot II - draws Silver
 Villager Bot II - reveals: Copper, Tactician, Watchtower
 Villager Bot II - places Watchtower on top of deck
 Villager Bot II - places Tactician on top of deck
 Villager Bot II - discards Copper
 Villager Bot II - plays Golem
 Villager Bot II - reveals Tactician, Watchtower
 Villager Bot II - plays Watchtower
 Villager Bot II - plays Tactician
 Villager Bot II - discards: Duchy, Estate, Copper, Estate, Estate, Tactician, Copper, Silver, Silver
 Villager Bot II - draws Estate, Estate
 Villager Bot II - shuffles deck Villager Bot II - draws Estate, Duchy, Copper

[Tombolo]

Villager Bot   cards: 20 Rats, 1 Estate

[Axxle]

Villager bot clearly just wants the attention.

[Warfreak2]

Yes this is a definite issue with the bots, also seen them frequently with 3-4 Golems but only 2-3 other actions in deck!!!

That's not particularly bad unless the other actions are some useless things, which of course is likely since these are bots, but still. If you have two Knights, or two Possessions, &c., you don't want any more non-Golem action cards because they will only dilute the effectiveness of your Golems. Golem/Counting House even wants only one non-Golem action card.

[Titandrake]

Yes this is a definite issue with the bots, also seen them frequently with 3-4 Golems but only 2-3 other actions in deck!!!

That's not particularly bad unless the other actions are some useless things, which of course is likely since these are bots, but still. If you have two Knights, or two Possessions, &c., you don't want any more non-Golem action cards because they will only dilute the effectiveness of your Golems. Golem/Counting House even wants only one non-Golem action card.

The meanest deck I have ever made was a game with no +Actions, but strong attacks. I got my deck down to 1 Ghost Ship, 1 Torturer, and lots of Golems.

It was absurdly evil.

[ponnuki]

It's a few days ago and I forgot the details, but it went something like this:

The bot's deck through whatever means had been reduced to Apprentice-Copper-2 or 3 Victory cards. The bot plays Apprentice, trashing a Copper, then buying a Copper. Next two turns the same, then I put it out of its misery by ending game.

[florrat]

In this game, Warlord Bot has found a new way to make a golden deck: http://dom.retrobox.eu/?/20130625/log.510876c7e4b0ac7a7a0bb212.1372176538499.txt.

The idea is to make your deck so small that you can play a Warehouse to discard all but one card of your hand. Then you can play a Bishop to get 1 VP and you don't even have to trash a card. So you get 1 VP every turn, and there's no need to green at all! How good is that?

He made one small mistake though, in my 17th turn. Then he trashed one of his estates. After that he had to discard his whole hand with his Warehouse. But at that moment he had already played 9 turns with his mega golden deck, so not much could go wrong anymore.

[Squidd]

plays Worker's Village
plays Laboratory
plays Laboratory
plays Worker's Village
plays Doctor
   names Curse
   reveals Silver
plays Doctor
   names Curse
   reveals Silver
plays Doctor
   names Curse
   reveals Silver

NB: There were no Curses in the deck, there was no curser in the game.

[A Drowned Kernel]

plays Worker's Village
plays Laboratory
plays Laboratory
plays Worker's Village
plays Doctor
   names Curse
   reveals Silver
plays Doctor
   names Curse
   reveals Silver
plays Doctor
   names Curse
   reveals Silver

NB: There were no Curses in the deck, there was no curser in the game.

Well obviously it didn't want to trash its silver, geez!

[clb]

Fairly classic Bot strategy - build your deck to the point where it starts coming together, then trash your hand down to Apprentice, Apprentice, Copper. Trash a Copper and buy a Copper each turn while your opponent feels shameful for having thought it would be a good game.
http://dom.retrobox.eu/?/20130703/log.50f6bcf3e4b0b0719c33eb27.1372913521422.txt

[Squidd]

Okay, V. Bot, you can continue buying Stonemason + double Hamlet over and over. I'll just be over here stockpiling Cities.

[Tombolo]

He had a funky little Advisor/Apprentice engine going, but it got kinda out of hand.  Here's his last turn:

Warlord Bot   plays Apprentice
Warlord Bot   trashes Copper
Warlord Bot   plays Apprentice
Warlord Bot   trashes Copper
Warlord Bot   buys Copper
Warlord Bot   gains Copper
Warlord Bot   shuffles deck
Warlord Bot   draws Apprentice, Copper, Apprentice

Yup, he trashed down to a three card deck.

[Tombolo]

...The bots surprise me sometimes.  I decided to see if I could get away with Contraband/FG, but the bot was smart enough to prohibit FG and I wound up shooting myself in the foot pretty badly!

[A Drowned Kernel]

I was playing through the intrigue adventure and discovered to my horror that the bots are smart enough to realize that scout and wishing well "combo" but not smart enough to realize that it's not actually a good idea, which leads to some pretty long and painful games even on the fastest speed. And then I saw one do this:

Soldier   plays Scout
Soldier   reveals: Masquerade, Ironworks, Silver, Copper
Soldier   places Silver on top of deck
Soldier   places Masquerade on top of deck
Soldier   places Ironworks on top of deck
Soldier   places Copper on top of deck
Soldier   plays Scout
Soldier   reveals: Copper, Ironworks, Masquerade, Silver
Soldier   places Masquerade on top of deck
Soldier   places Silver on top of deck
Soldier   places Ironworks on top of deck
Soldier   places Copper on top of deck

It not only plays scout twice in a row to reveal the exact same four cards, but it puts them back in a different order the second time. I guess it changed its mind about what it wanted to draw next.

(Turn 19)

[clb]

This is totally dead horse (http://qkme.me/3vlvoh), but the Rats problem with the Bots is out of control. How hard would it be to give each Bot 2 lines of code: 1 that says do not ever buy a Rats and 2 that says do not ever play a Rats. If you wanted to get fancy, you could allow the Bots to trash Rats, but that might be more work than it's worth.

edit: most recent suffering of Bot-Rats (http://dominionlogs.goko.com//20130826/log.50f6bcf3e4b0b0719c33eb27.1377560408462.txt) (sorry, not pretified since I am at work).

[Krarks_pinky]

Bot playbook:

1. Open young witch no matter what. Get 2 or 3 or even more.
2. Cursers are better than anything else in a kingdom.
3. Play as many warehouses as you can each turn, maybe your opponent will quit from boredom.
4. Open potion as a top priority, even if it's only to get a golem or philosopher stone.
5. Piles getting a bit low? Better start greening, even if it's just a duchy.
6. Trash for benefit / remodeling is probably pretty good, just go for it and see what happens.

[Wrclass]

Bot playbook:

1. Open young witch no matter what. Get 2 or 3 or even more.
2. Cursers are better than anything else in a kingdom.
3. Play as many warehouses as you can each turn, maybe your opponent will quit from boredom.
4. Open potion as a top priority, even if it's only to get a golem or philosopher stone.
5. Piles getting a bit low? Better start greening, even if it's just a duchy.
6. Trash for benefit / remodeling is probably pretty good, just go for it and see what happens.

I have only the base set, but I definitely noticed 2, 5 and 6

[DG]

The duchy purchasing strategy for the bots is based on 3 or 4 player games. Buying duchies when one pile is empty seems much more reasonable in that setting. The bots also follow very similar purchasing rules so together they can empty piles making early duchies more valuable.

[clb]

The duchy purchasing strategy for the bots is based on 3 or 4 player games. Buying duchies when one pile is empty seems much more reasonable in that setting. The bots also follow very similar purchasing rules so together they can empty piles making early duchies more valuable.

I have been the victim of Bot-group-think. If everyone is doing the same bad idea, your idea has to be a LOT better or else you're better off just rolling the dice and hoping that you can do their bad idea better (luckier).

[ragingduckd]

How hard would it be to give each Bot 2 lines of code: 1 that says do not ever buy a Rats and 2 that says do not ever play a Rats.

Judging by this month's Goko "update," it's probably about 3-4 days hard.

[Beyond Awesome]

The duchy purchasing strategy for the bots is based on 3 or 4 player games. Buying duchies when one pile is empty seems much more reasonable in that setting. The bots also follow very similar purchasing rules so together they can empty piles making early duchies more valuable.

I have been the victim of Bot-group-think. If everyone is doing the same bad idea, your idea has to be a LOT better or else you're better off just rolling the dice and hoping that you can do their bad idea better (luckier).

Yah, it's called Adventures.

[pst]

In this game Lord Bottington bought four potions! The only Potion cost card was Transmute, so we wondered what he needed all those Potions for. (He never got any Transmutes at all by the way, or used the Potions in any other way.)

[florrat]

Wow, this might have been the worst strategy a bot has played against me:

http://dom.retrobox.eu/?/20131019/log.510876c7e4b0ac7a7a0bb212.1382239938990.txt

- Buying 5 Conspirators, without having a single non-terminal card at any time during the game (except Ruined Villages)
- Buying Duchy on the first and second time it reached $5 (turn 6 and 8, while Cultist was on the board)
- Buying first Cultist while there were just 2 Ruins left
- Buying Potion on a board with only Transmute, but not on turn 1 or 2.

[Puddleglumm]

A few days ago Warlord Bot constructed this fantastic, exploding, do-nothing engine out of Universities, Ghost Ships, Advisors, Warehouses, and Moats. (Well, one thing it was doing was Ghost Shipping me every turn) It was always discarding coppers with it's Warehouses, so it could never afford more than a University, and each turn it was gaining multiple "engine" parts with it's existing universities. I present to you Turn 20, in which Warlord Bot plays 28 actions, gains 1 Ghost ship and 5 copies of Traders via University, apparently setting of it's internal piledriving alarm along the way, decides it's time to start greening and ... buys an estate. Maybe next turn it was going for the coveted Trader Megaturn and was going to Trader 5 Ghost ships into 25 silver?
 
Warlord Bot: turn 20
----------
Warlord Bot - plays Advisor
Warlord Bot - reveals: Moat, Warehouse, Copper
Puddleglumm - discards Moat
Warlord Bot - places cards in hand: Warehouse, Copper
Warlord Bot - plays Advisor
Warlord Bot - reveals: Warehouse, University, Warehouse
Puddleglumm - discards Warehouse
Warlord Bot - places cards in hand: Warehouse, University
Warlord Bot - plays University
Warlord Bot - gains Ghost Ship
Warlord Bot - plays Ghost Ship
Puddleglumm - reveals reaction Moat
Warlord Bot - draws Advisor, Warehouse
Warlord Bot - plays Warehouse
Warlord Bot - draws Peddler, University, Ghost Ship
Warlord Bot - discards Estate Warlord Bot - discards Estate
Warlord Bot - discards Copper
Warlord Bot - plays University
Warlord Bot - gains Trader
Warlord Bot - plays Advisor
Warlord Bot - reveals: Warehouse, Advisor, University
Puddleglumm - discards Warehouse
Warlord Bot - places cards in hand: Advisor, University
Warlord Bot - plays University
Warlord Bot - gains Trader
Warlord Bot - plays Advisor
Warlord Bot - reveals: Advisor, Advisor, Copper
Puddleglumm - discards Advisor
Warlord Bot - places cards in hand: Advisor, Copper
Warlord Bot - plays Advisor
Warlord Bot - reveals: Copper, Advisor, Copper
Puddleglumm - discards Advisor
Warlord Bot - places cards in hand: Copper, Copper
Warlord Bot - plays Peddler
Warlord Bot - draws Ghost Ship
Warlord Bot - plays Ghost Ship
Puddleglumm - reveals reaction Moat
Warlord Bot - draws Warehouse, Copper
Warlord Bot - plays Ghost Ship
Puddleglumm - reveals reaction Moat
Warlord Bot - draws Ghost Ship, Potion
Warlord Bot - plays Warehouse
Warlord Bot - draws Moat, University, Estate
Warlord Bot - discards Estate
Warlord Bot - discards Copper
Warlord Bot - discards Copper
Warlord Bot - plays University
Warlord Bot - gains Trader
Warlord Bot - plays Ghost Ship
Puddleglumm - reveals reaction Moat
Warlord Bot - draws University, Estate
Warlord Bot - plays University
Warlord Bot - gains Trader
Warlord Bot - plays Warehouse
Warlord Bot - draws Warehouse, Copper, Copper
Warlord Bot - discards Estate
Warlord Bot - discards Copper
Warlord Bot - discards Copper
Warlord Bot - plays Moat
Warlord Bot - draws University, Ghost Ship
Warlord Bot - plays University
Warlord Bot - gains Trader
Warlord Bot - plays Ghost Ship
Puddleglumm - reveals reaction Moat
Warlord Bot - draws Estate, Ghost Ship
Warlord Bot - plays Warehouse
Warlord Bot - draws Copper, Warehouse
Warlord Bot - shuffles deck
Warlord Bot - draws Copper
Warlord Bot - discards Estate
Warlord Bot - discards Copper
Warlord Bot - discards Copper
Warlord Bot - plays Warehouse
Warlord Bot - draws Estate, Trader, Moat
Warlord Bot - discards Estate
Warlord Bot - discards Moat
Warlord Bot - discards Trader
Warlord Bot - plays Warehouse
Warlord Bot - draws Estate, Warehouse, Trader
Warlord Bot - discards Estate
Warlord Bot - discards Trader
Warlord Bot - discards Copper
Warlord Bot - plays Warehouse
Warlord Bot - draws Copper, Advisor, Advisor
Warlord Bot - discards Copper
Warlord Bot - discards Copper
Warlord Bot - discards Potion
Warlord Bot - plays Advisor
Warlord Bot - reveals: Copper, Trader, Copper
Puddleglumm - discards Trader
Warlord Bot - places cards in hand: Copper, Copper
Warlord Bot - plays Advisor
Warlord Bot - reveals: Estate, Estate, Trader
Puddleglumm - discards Trader
Warlord Bot - places cards in hand: Estate, Estate
Warlord Bot - plays Ghost Ship
Puddleglumm - reveals reaction Moat
Warlord Bot - draws Warehouse, Ghost Ship
Warlord Bot - plays 2 Copper
Warlord Bot - buys Estate
Warlord Bot - gains Estate
Warlord Bot - draws Trader, Copper
Warlord Bot - shuffles deck
Warlord Bot - draws Advisor, Copper, Advisor
 
 
Embarassingly, I was more or less buying the same cards to no useful end, just with enough common sense to get some treasure here and there, which enabled me to snag a few Dutchies when I saw where Warlord Bot's engine was heading.
 

[Puddleglumm]

Bot playbook:
3. Play as many warehouses as you can each turn, maybe your opponent will quit from boredom.
4. Open potion as a top priority, even if it's only to get a golem or philosopher stone.

These two crack me up! Once I saw a bot randomly buy a potion in the middle of the game so it could pick up an apothecary. I guess bots love useless cantrips so much they're willing to acquire potions mid-game to get them. 

[eHalcyon]

Bot playbook:
3. Play as many warehouses as you can each turn, maybe your opponent will quit from boredom.
4. Open potion as a top priority, even if it's only to get a golem or philosopher stone.

These two crack me up! Once I saw a bot randomly buy a potion in the middle of the game so it could pick up an apothecary. I guess bots love useless cantrips so much they're willing to acquire potions mid-game to get them. 

Apothecary is far from useless, but getting them mid-game usually isn't a great idea.

[Puddleglumm]

Bot playbook:
3. Play as many warehouses as you can each turn, maybe your opponent will quit from boredom.
4. Open potion as a top priority, even if it's only to get a golem or philosopher stone.

These two crack me up! Once I saw a bot randomly buy a potion in the middle of the game so it could pick up an apothecary. I guess bots love useless cantrips so much they're willing to acquire potions mid-game to get them. 

Apothecary is far from useless, but getting them mid-game usually isn't a great idea.

Of course you're correct, I was more trying to point out the ridiculousness of buying your first potion mid-game just to get an apothecary- the deck is already thick, and the apothecary it would buy can only come 2 shuffles later, at which point hitting even 1 copper with it would probably be a nice outcome.

[Schneau]

The Goko bots go bonkers when stonemason and cantrips are on the board. Occasionally they will pull a nifty 3-pile ending, using the stonemasons to trash for VP at the last minute, but often they will just build a deck full of lighthouses and spies. I recommend "Stonemason, All" against bots for your entertainment needs. 

I took this advice from another thread, and forced Stonemason and Rats, with the rest random. The remaining cards created a kind of Bot Paradise, with many favorites such as Warehouse, Oasis, and Alchemist. So, after Villager Bot opened Silver/Potion/Potion, I Embargoed Alchemist, and VB decided to switch strategies into mass Warehouse. I was a bit disappointed VB never picked up a Rats, but it still managed to rarely get even $5.

[Puddleglumm]

The Goko bots go bonkers when stonemason and cantrips are on the board. Occasionally they will pull a nifty 3-pile ending, using the stonemasons to trash for VP at the last minute, but often they will just build a deck full of lighthouses and spies. I recommend "Stonemason, All" against bots for your entertainment needs. 

I took this advice from another thread, and forced Stonemason and Rats, with the rest random. The remaining cards created a kind of Bot Paradise, with many favorites such as Warehouse, Oasis, and Alchemist. So, after Villager Bot opened Silver/Potion/Potion, I Embargoed Alchemist, and VB decided to switch strategies into mass Warehouse. I was a bit disappointed VB never picked up a Rats, but it still managed to rarely get even $5.

Villager Bot   overpays for Stonemason with 3 coins
Villager Bot   gains Warehouse
Villager Bot   gains Warehouse
Villager Bot   gains Stonemason

This is the quintessential villager bot play! 

[Morgrim7]

He had a funky little Advisor/Apprentice engine going, but it got kinda out of hand.  Here's his last turn:

Warlord Bot   plays Apprentice
Warlord Bot   trashes Copper
Warlord Bot   plays Apprentice
Warlord Bot   trashes Copper
Warlord Bot   buys Copper
Warlord Bot   gains Copper
Warlord Bot   shuffles deck
Warlord Bot   draws Apprentice, Copper, Apprentice

Yup, he trashed down to a three card deck.

I think this may be getting to a point where one might consider Ambassadoring an Apprentice to the bots...