Print

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

[Infthitbox]

I'm having trouble determining what I should take away from reviewing my game logs, particularly when I think my plan was OK and then I get smashed, or when it seems like there is an obvious reason why I lost, but it may hide (or even be caused by) earlier mistakes. I also have trouble reviewing logs from victories, because the question: "How could I have lost to a better player if I play this way?" is extremely difficult to answer (because I'm not a better player than myself). I'm going to post 4 game logs here, with the questions I asked myself and what I think I should learn from the game. The idea here is primarily getting better at analysis, but specific insight into the actual mistakes made would be appreciated as well.

Code: [Select]

Duchess, Herbalist, Native Village, University, Coppersmith, Contraband, Counterfeit, Mint, Vault, Nobles
http://dominion-game-logs.s3.amazonaws.com/game_logs/20151115/log.0.1447604212525.txt

I look at this board and decide I want to draw my deck each turn with Native Village and Nobles. I open Silver/Silver, get a Counterfeit, and start acquiring Nobles. My opponent plays Big Money with Vault. Its tempting to just look at the leaderboard and conclude that Vault-BM was the correct strategy here and move on, but that doesn't actually help me here. I don't think I built my deck properly; it feels like a Vault would have been good, both early to hit 6 and then later when I can discard some cards and redraw them with Nobles afterwards. I feel like I waited too long to pick up a second source of +buy, which felt like my only way to get back in the game after I got into a VP hole. I asked myself whether the engine approach here is worthwhile, and I don't see great payload (as an aside, as someone new to "organized strategy" but who has played for years at a "pretty good for someone who never thought too hard about it" level, I find the terminology difficult to adapt to, as I had phrases mentally for some of these concepts), so its possible that its not worth it and that Vault-BM is too good. There isn't any variable-VP either, which means that a Nobles advantage will only get me so far. This one is honestly hard; I really want there to be something other than Vault-BM here.

Code: [Select]

Fool's Gold, Ambassador, Horse Traders, Throne Room, Walled Village, Catacombs, Jester, Mine, Mountebank, Saboteur

http://dominion-game-logs.s3.amazonaws.com/game_logs/20151115/log.0.1447603480865.txt


I open 5/2 here, which is possibly the conclusion. It doesn't necessarily feel that way to me (although Rabid seemed to think so in chat, possibly he was trying to be nice). At about turn 8/9, I felt like I was in the game. I hit 5 and buy a Catacombs, which for several "hand-occurrences" feels incorrect to play, so I don't. Rabid picks up a Throne Room at this time, which I hadn't really considered and didn't until it was far, far too late. I also picked up a Saboteur when I felt I was so far behind that I needed a miracle, but it was way too late by that point as well. Again, its tempting to be reductive and say that the opening killed me here, but that doesn't seem right for a game where it felt like with better buys turns 9-11 or so that I could have been in it.

This was the first of the two games I played against Rabid, the first log presented here being the second, but I swapped the orders of them for the post because I feel like the first log is more interesting as a review.

Code: [Select]

Haven, Pawn, Squire, Scrying Pool, Masquerade, Tunnel, Advisor, Counterfeit, Festival, Rogue

http://dominion-game-logs.s3.amazonaws.com/game_logs/20151116/log.0.1447643839204.txt

I win this game when my opponent concedes when I "get so far ahead" according to his chat. I open Masquerade/Potion and pick up Advisor/Scrying Pool on 3/4. Its hard for me to gauge the effectiveness of the Advisor here, as the first time playing it I get denied a Potion (opponent's choice over Masquerade, not sure there, felt correct to me), but that motivates me to pick up a Counterfeit, which turns out to be great (predictably). I start getting Festivals and Pawns, but I also buy a second Advisor, which was terrible. Shortly thereafter, I have enough Scrying Pools where its irrelevant to play Advisor and I feel like it was probably better off just being a Pawn. My opponent concedes after I Counterfeit my Potion to pick up the last to Scrying Pools; I actually prefer to continue playing, as I really need the practice on pile control and actually winning the game from that spot without bricking my deck with VP. The questions here involve the Advisor buys, and whether or not the game actually swung on my opponent buying Silver and Squire and polluting his deck. It feels like the second Advisor was wrong, but I'm not sure on that first one, would Pawn have been better? I'm also not sure whether the game actually hinged on discarding his potion with Scrying Pool on turn 6 and his own Scrying Pool on turn 8. That feels like a big source of my subsequent advantage, but that's difficult to gauge.

Code: [Select]

University, Masquerade, Tunnel, Philosopher's Stone, Bridge, Farming Village, Monument, Taxman, Merchant Ship, Vault

http://dominion-game-logs.s3.amazonaws.com/game_logs/20151116/log.0.1447649466137.txt

This game bears some similarity to the first log. I try to play an engine, which is almost certainly overrun with terminals. My opponent plays Taxman-BM, and crushes me. This one feels more like a misbuilt deck; although it could be Vault/Masquerade just aren't good enough at card drawing to get the job done. I also wonder whether I should have picked up a Monument earlier than I did, I knew pretty early that I needed a lot of VP chips to beat an unfavorable Province split. Between this game and the first log, it really looks like l need to improve my ability to judge the available engine strength versus the relevant BM strategy. This one feels particularly bad to me, as surely Vault-BM is better than Taxman-BM? It also feels like between Monument and Bridge that something other than Big Money was viable here. That or I just don't understand the power of Vault.




What types of questions ought I to be asking myself after these kinds of games? Is my analysis way off base, or is it on the right track and I need to improve what I think I need to improve? Concluding that I need to work on something when that wasn't the primary (or secondary, etc) takeaway from a game leads to wasted time.

[ehunt]

Game one, I would have gone big money/vault as well. I saw the board before I read your story and I thought "nothing to do, vault/big money." But I'll bet some of the more engine oriented folks will go for the engine here, and I do think it's reasonably close (vault/big money helps the engine because of vault's discard clause).

I guess the place where we fundamentally disagree: I just think Native Village/Smithy is not very good. Even Village/Smithy, without a gainer around*, is not very good, and Native Village is substantially worse than Village. Native Village/Nobles is better and worse: better because the points on the Nobles  help to make up for the time it takes to buy them and because occasionally you'll start a hand with two Nobles; worse because Nobles are so expensive.
*yes, there's University, but since it can't gain Nobles and I am just talking about Smithy for the purpose of analogy, I ignore it here.

Now's where the "no payload" argument really gets strong: against Vault/Big Money, there's just nothing the engine can do to slow the opponent down. With a discard attack, it would be safe to build the engine till the opponent had as many as 4 provinces: once you get your discard attack going every turn, your opponent will be unable to buy a province ever again or even a duchy most turns, and your nobles advantage will give you an easy win. Even without a discard attack, but with more reliable +buy and better money generation, you could probably let the opponent get as many as 3 provinces before you started worrying about greening. But here, there's just no time. The (unattacked) vault deck is the energizer bunny: it just keeps going. It has to draw a vault somewhere near a gold, and that's it. Everytime it draws a vault without a gold, it can buy a gold to make it that much more likely that it won't whiff the next shuffle.

[DG]

Game 1 - By turn 4 you've bought three silver and a counterfeit. That's always going to be trouble when making an action engine.
Game 2 - Increasing income to get out of an ambassador battle is important and fool's gold was better than silver here.
Game 3 - As you trash your starting cards from your deck you should be able to draw the deck well without the advisors. Step up to festivals and rogues sooner. Trashing squires to gain attacks was also a thing here and it happens by buying a second masquerade.
Game 4 - It looks possible to make an engine here but drawing cards is the difficulty. Again, get a second masquerade early to trash down. Use the universities to gain farming villages and vaults then add either monuments or bridges (not both) once you have an engine running. Universities don't draw cards so don't get too many (or be willing to trash some).

Something to remember about vaults is that they can sustain a treasure strategy when big money would usually slow down (after 4 provinces).

[Dingan]

Regarding game 1...
When I see typical BM strats (Vault, Embassy, Jack, etc.), I look for a way to slow them down.  If there is no way, which is the case on your board, the BM strat will do its thing, uninterrupted.

I then look for a way to surpass it.  I don't really see that here.  Maybe you could get fancy with Minting Golds, and Counterfeit for +buy, then double-green in while your opponent is only single greening.  But I think that would be a bit too slow to set up.

[Limetime]

For game 1: Game 1 you should either do big money vault, Draw engine with coppersmith as payload, or native village/ coppersmith with herbalist as + buy.
For game 2: If your feeling foolish then open mine/fool's gold. Opening mountebank is ok because the economy you get from them.

[assemble_me]

Draw engine with coppersmith as payload, or native village/ coppersmith with herbalist as + buy.

I don't see that. The problem for the engine is that the villages don't draw, so you only increase your handsize by 1 after playing a village and a Nobles.
For Coppersmith you need a reliable deck, good villages or really big draw. So treasures seem the way to go, getting a Mint to get thin and later to mint Gold sounds better than Coppersmith to me, here. Also, Counterfeit is just much better so you don't need Herbalists.
But still, BM-Vault is really good (Counterfeit helps as well) and will do its thing.

[luser]

Game 1, vault is thing here, you could also trash 4+ coppers with mint when you already have vault for more reliability.
Game 2, Open mountebank/fools gold ambassador could wait until second shuffle.
Game 3, Here get second masq and you forgot about star of this board, squire. You should always trash squire with masq, gain scrying pool.
Game 4 is obvious vault-tunnel. Trying to make moat-village engine does not work unless you have strong trashing. If there was better draw then taxman is one of best attacks versus bm, of course you need to use him for gold->gold tax.

[Infthitbox]

Now's where the "no payload" argument really gets strong: against Vault/Big Money, there's just nothing the engine can do to slow the opponent down. With a discard attack, it would be safe to build the engine till the opponent had as many as 4 provinces: once you get your discard attack going every turn, your opponent will be unable to buy a province ever again or even a duchy most turns, and your nobles advantage will give you an easy win. Even without a discard attack, but with more reliable +buy and better money generation, you could probably let the opponent get as many as 3 provinces before you started worrying about greening. But here, there's just no time. The (unattacked) vault deck is the energizer bunny: it just keeps going. It has to draw a vault somewhere near a gold, and that's it. Everytime it draws a vault without a gold, it can buy a gold to make it that much more likely that it won't whiff the next shuffle.

Regarding game 1...
When I see typical BM strats (Vault, Embassy, Jack, etc.), I look for a way to slow them down.  If there is no way, which is the case on your board, the BM strat will do its thing, uninterrupted.

I then look for a way to surpass it.  I don't really see that here.  Maybe you could get fancy with Minting Golds, and Counterfeit for +buy, then double-green in while your opponent is only single greening.  But I think that would be a bit too slow to set up.

Comparing two (or more) disparate strategies at the start of the game is something I struggle with, so good heuristics like this are very helpful.

For game 2: If your feeling foolish then open mine/fool's gold. Opening mountebank is ok because the economy you get from them.

Game 2, Open mountebank/fools gold ambassador could wait until second shuffle.

I was scared of Mountebank here when I suspected my opponent would have two ambassadors and that I would be late to the party getting even one.

Game 3, Here get second masq and you forgot about star of this board, squire. You should always trash squire with masq, gain scrying pool.

I always forget Squire's "below the fold" ability. This would have been good. I was too concerned with acquiring cheap +buy from Pawn.

Game 4 is obvious vault-tunnel. Trying to make moat-village engine does not work unless you have strong trashing. If there was better draw then taxman is one of best attacks versus bm, of course you need to use him for gold->gold tax.

[faust]

Draw engine with coppersmith as payload, or native village/ coppersmith with herbalist as + buy.

I don't see that. The problem for the engine is that the villages don't draw, so you only increase your handsize by 1 after playing a village and a Nobles.
For Coppersmith you need a reliable deck, good villages or really big draw. So treasures seem the way to go, getting a Mint to get thin and later to mint Gold sounds better than Coppersmith to me, here. Also, Counterfeit is just much better so you don't need Herbalists.
But still, BM-Vault is really good (Counterfeit helps as well) and will do its thing.

It isnt' correct to say that Native Village doesn't draw. Playing n Native Villages only nets you one card less than playing n Villages (assuming you don't need the drawn cards in between).

I think engine is correct. I would try to make Mint/Counterfeit work.

[Limetime]

Draw engine with coppersmith as payload, or native village/ coppersmith with herbalist as + buy.

I don't see that. The problem for the engine is that the villages don't draw, so you only increase your handsize by 1 after playing a village and a Nobles.
For Coppersmith you need a reliable deck, good villages or really big draw. So treasures seem the way to go, getting a Mint to get thin and later to mint Gold sounds better than Coppersmith to me, here. Also, Counterfeit is just much better so you don't need Herbalists.
But still, BM-Vault is really good (Counterfeit helps as well) and will do its thing.

For the coppersmith strategy I meant a megaturn strategy.

[Dingan]

Comparing two (or more) disparate strategies at the start of the game is something I struggle with

We all struggle with this.  In fact, it's one of my favorite things about Dominion.  When I see 2 or more disparate strategies, strats A and B, I will pick either A or B.  Then if my opponent chooses the other one, I will say "well, we'll see which one is better".

Sometimes it's really close, and it comes down to a dud turn, or having $7 instead of $8 on some turn in the end game, or something dumb and trivial.  In which case, you can pretty much either say they're both equally good, or that you would need to play more games with that Kingdom to determine which one is good.

But other times, it's completely clear by the middle of the game that one is so incredibly obviously the better one, which I see in hindsight, but hadn't seen at the beginning of the game.  I feel like an idiot (whether I'm winning or losing).  But I don't get "frustrated" by these moments.  Rather, I see them as tremendous learning opportunities, and I make a mental note of what went down and what I should takeaway from the game.

[Infthitbox]

Comparing two (or more) disparate strategies at the start of the game is something I struggle with

We all struggle with this.  In fact, it's one of my favorite things about Dominion.  When I see 2 or more disparate strategies, strats A and B, I will pick either A or B.  Then if my opponent chooses the other one, I will say "well, we'll see which one is better".

Sometimes it's really close, and it comes down to a dud turn, or having $7 instead of $8 on some turn in the end game, or something dumb and trivial.  In which case, you can pretty much either say they're both equally good, or that you would need to play more games with that Kingdom to determine which one is good.

But other times, it's completely clear by the middle of the game that one is so incredibly obviously the better one, which I see in hindsight, but hadn't seen at the beginning of the game.  I feel like an idiot (whether I'm winning or losing).  But I don't get "frustrated" by these moments.  Rather, I see them as tremendous learning opportunities, and I make a mental note of what went down and what I should takeaway from the game.

The problem is that I don't always know if I chose the wrong strategy (see game 1 where most people agree Vault-BM is better than what I tried) or if I played the "better" strategy wrong: if I had the right general idea but bought my cards in the wrong order, or failed to identify an important piece (the Squire-Scrying Pool-trashing interaction from game 3), or if I just got unlucky enough early that it handicapped my strategy enough that it looked worse at the end of the game to make me draw the wrong conclusion. Making a mental note of what went wrong (or right) is ideally what everyone wants for every game, but even evaluating what that takeaway should be is very difficult, which I was trying to drive at in this topic.

Some games, on the other hand, it is obvious (even to me) that I chose the wrong strategy. But in most games, I don't put enough faith in my ability to play my chosen strategy well enough to think it is worth drawing conclusions on. Hence, game reports forum.

[dedicateddan]

Late to the party, but:

From game logs, I like to look at:
1) What's the most powerful strategy on the board? Is there a powerful engine? Is there alt-VP or an attack to punish non-engine decks?
2) What's the baseline strategy for the board? How much trashing do you want? How much do you build before greening? Playing the board a couple of times helps with this
3) What variations from the baseline strategy are needed? (openings, draws, piles, points, etc.)

Board 1: There's no attack to punish BM, so Vault/Nobles/Counterfeit BM is going to be faster than native village + nobles
Board 2: Ambassador war, whoever wins has complete control
Board 3: Scrying pool + mass actions -> megaturn board. You want to get thin, so I like double masquerade
Board 4: Depending on how the game play out, you can threaten monument point tokens, a bridge megaturn, or both. Vault tunnel is quite strong. You might not have time for universities

[Infthitbox]

Late to the party, but:

From game logs, I like to look at:
1) What's the most powerful strategy on the board? Is there a powerful engine? Is there alt-VP or an attack to punish non-engine decks?
2) What's the baseline strategy for the board? How much trashing do you want? How much do you build before greening? Playing the board a couple of times helps with this
3) What variations from the baseline strategy are needed? (openings, draws, piles, points, etc.)

By "baseline strategy" do you mean the building blocks of the best strategy, or do you mean a simple, known quantity to measure any other plans against?

Mentally, I've been using "baseline strategy" to refer to one or two card strategies that I need to judge any other plans against. Vault-BM (only after the games in this thread, when I found out firsthand), Envoy-BM, Hermit-Market Square, etc. Of course, I only have a very rough idea of how to compare a "known quantity" to my plan, because I haven't had a ton of relevant experience to know.

[dedicateddan]

I'm referring to what you're generally trying to do on a board, although sometimes you might deviate.

For example, a board might favor engine, but if you open 5-2, you might consider going for vault big money.