Print

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

[philosophyguy]

At what point in the game do you decide that your best (only) chance to win is with a high-risk, high-reward strategy? Specific examples will of course depend on the setup, but some of the thresholds you might consider are getting 2P after your first shuffle with Familiar, having your Opponent's Sea Hag flip your Sea Hag on turn 3, or ending on the bad end of a race for a pile (e.g., Minion). The kind of "I've got nothing to lose" strategy I have in mind might be Treasure Map without support.

[Death to Sea Hags]

Well, in that case, unsupported Treasure Map is too slow to pay off.  It's a risky-enough strategy with 10 other cards in your deck.  Trying it with 20+ is NOT going to hit.  As for other high-variance strategies, the same problem exists - You've got to buy it, then draw it - and by the time you want to try them, it's usually too late to make them work.

Other than TM, what else could you do....
Late game Sab to hit a Province?  Your opp has enough $3+ cards to shield the Prov, unless you've got Rabble or Fortune Tellers.
Tournament isn't much of a late-game strat because its played out by then.
Mass Swindler might be worth it if you have actions.
Forge? if you have enough +draw, it's okay.
Tactician? very streaky.
Black Market for a single power card?

[Kirian]

The only real nothing-to-lose strategy I can think of is breaking the PPR.  If your opponent has a reasonable chance of not drawing $8 in any given round--e.g. they're not drawing their deck, not playing at ultra-dense money--then if you're 10 points behind, it can be worth it to go for broke.  Otherwise... Saboteur, perhaps, or Tactician for a single mega-turn.

[Geronimoo]

In the long run it's probably better to not depart from optimal strategy even if you got a little unlucky in the early turns.

[Superdad]

This is a strategy that people probably aren't even used to yet, as thsi game is still young(ish). For example, this line of strategic thinking has existed in MtG for a LONG time, but it also took a LONG time for people to understand it.

For example, in mtg, if you are far enough behind that the only possible chance you are winning is if X card is on top of your deck, then you PLAY as though it is. Similarly, if you will 100% lose if your opponent has a certain card, then you play as though they don't have it.

If you want a great read on the topic, take a look at point #7 in the following article:  http://www.starcitygames.com/magic/fundamentals/12632_Better_Lucky_Than_Good_Seven_Ways_to_Get_Lucky.html

There have been some great examples of this, such as Craig Jones' $16,000 Lightning Helix topdeck:

Video: http://www.google.ca/url?sa=t&source=web&cd=2&ved=0CB8QtwIwAQ&url=http%3A%2F%2F
Article: http://www.starcitygames.com/magic/standard/11615_The_16000_Lightning_Helix_A_PT_Honolulu_Report_Part_2_2nd.html
 (take a look at the semifinals match against Oliver Ruel)

Basically, he was dead, almost no chance to win the game. He formulated a plan that could possibly win, but only if a certain combination of cards came up. Since he was going to lose regardless, he played for the plan as though it would happen. And it did. It would not have won if he didn't plan and play for it many turns in advance. That last lightning helix would not have won the game if he didn't make seemingly sub-par decisions before-hand, but he knew it was his only out, so he played for it.

Truly great dominion players will start to pick this up. If you are so far behind that the only way you are winning the game is if a certainly thing happens, then play for that certain thing to happen.


A dominion example:

- There are two piles empty, and no other options for VPs, besides province/duchy/estate (i.e. no goons, bishop, gardens, etc).
- There are 2 provinces and 2 duchies remaining.
- You are behind by 8 points.
- You have $7 on your turn.

The only way you are going to win this game is if you get both provinces and your opponent gets no more than 1 duchy, 1 estate.
The only way you are going to win this game is if you buy an ESTATE with your $7, then buy back to back provinces, while the opponent only buys a single duchy and a single estate.

This is a great example of "you 100% lose if your opponent has a certain card, so you MUST play as though they don't have it". You WILL 100% lose if your opponent can double-duchy, or if they can single-province in the next 2 turns. So your only option is to formulate a plan where they cannot do so (since you lose if they can).

So, if that's your only chance of winning, then execute the plan as though they can't buy a province or two duchies. If you buy the duchy with your $7, he ends on piles next turn. So buy the estate, and hope for the double $8 on your next 2 turns (provided it's possible), and hope he can't double duchy or single province. If it's literally your only shot, then play for it.

You might just turn up that $16,000 lightning helix off the top of the deck for the win.

[tko]

This is a strategy that people probably aren't even used to yet, as thsi game is still young(ish). For example, this line of strategic thinking has existed in MtG for a LONG time, but it also took a LONG time for people to understand it.

For example, in mtg, if you are far enough behind that the only possible chance you are winning is if X card is on top of your deck, then you PLAY as though it is. If you want a great read on the topic, take a look at:  http://www.starcitygames.com/magic/fundamentals/12632_Better_Lucky_Than_Good_Seven_Ways_to_Get_Lucky.html

I was fortunate enough to learn how to play Magic at a card shop where Pat Chapin (author of that article) used to play.  He's a genius at jedi mind tricking his opponents into playing into his strategy, and one of my favorite quotes from him regarding my deck with 3 Necropotences (when I should have played the full 4)... "Would you drive a car with only 3 tires?"

[Superdad]

That's awesome. I would have loved to play in Michigan, just to play with him. My playgroup calls (called) me the Canadian Chapin. So many times over the last 20-odd years, we'd be discussing a strategy in our play group, then Chapin would write an article about it a few weeks later. It was eerie. Him and Steven Menendian are my favorite players of all time. They just understand parts of the game that few people do. They are the type of player that picks up a game that has existed for a decade and completely break it open - and cause people that have been playing it to realize they didn't even understand the game that they've been playing for 10 years.

[Davio]

I sometimes try to catch my opponents off guard by trying to sneakily empty some 3rd pile + 1 Estate (or empty the Estates) if I have lost a GM battle or something like that. It's the Piledriver achievement on Councilroom.

This only works against lower ranked opponents, better ones may take note of this and buy just that single Colony to thwart your plan.

[philosophyguy]

This is a strategy that people probably aren't even used to yet, as thsi game is still young(ish). For example, this line of strategic thinking has existed in MtG for a LONG time, but it also took a LONG time for people to understand it.

For example, in mtg, if you are far enough behind that the only possible chance you are winning is if X card is on top of your deck, then you PLAY as though it is. Similarly, if you will 100% lose if your opponent has a certain card, then you play as though they don't have it.

This is exactly the kind of thinking I had in mind with the original question. What I'm wondering is, at what point do you decide that you are so far behind that you go for broke? I think a good player can probably make that judgment earlier than in your 2 Province / 2 Duchy example, but I don't know how much earlier. Is it too early if your Sea Hag gets flipped by your opponent's Sea Hag during turns 3-4?

[DG]

I'll certainly consider alternative strategies when looking at a losing deck. This earliest time might be when playing the same strategy as an opponent but they've grabbed the key cards, so following the same strategy with a worse deck would always be likely to fail. The easiest thing to change is the vp purchasing strategy, hoping to find any advantage from shortening or lengthening the game. Buying different attack cards might help but quite often an opponent can play through the one or two times they'll hit.

Here's a game showing a change of plan that actually did work http://councilroom.com/game?game_id=game-20110601-164555-8365414e.html.

[Glooble]

I just played an alchemist game where my opponent thoroughly trounced me in the alchemist race - 7 to 3 by the end. So I started buying outposts, figuring that with alchemist those extra turns would be more useful. Then I embargoed the outposts to keep him from following me. It worked - two of my outpost turns gave me provinces which let me catch up to his lead. That was a late game strategy shift that required a certain amount of luck to work out, since even my mini hands kept turning potions and keepng my alchemists on top, an outcome that was pretty unlikely.

[WanderingWinder]

I'll certainly consider alternative strategies when looking at a losing deck. This earliest time might be when playing the same strategy as an opponent but they've grabbed the key cards, so following the same strategy with a worse deck would always be likely to fail. The easiest thing to change is the vp purchasing strategy, hoping to find any advantage from shortening or lengthening the game. Buying different attack cards might help but quite often an opponent can play through the one or two times they'll hit.

Here's a game showing a change of plan that actually did work http://councilroom.com/game?game_id=game-20110601-164555-8365414e.html.

Certainly a thought to have. But on the other hand, you should also consider that if they've outraced you to the key cards, they'll probably beat you - but it might be even worse if you switch strategies. It depends both on how close the alternative strategies are, and how timing-dependent the different strategies are.

[philosophyguy]

But on the other hand, you should also consider that if they've outraced you to the key cards, they'll probably beat you - but it might be even worse if you switch strategies. It depends both on how close the alternative strategies are, and how timing-dependent the different strategies are.

This is entirely true but misses the point. Dominion is a win/lose system; there's no points for margin of victory/loss. So, if you're in a situation in which your chosen strategy is likely to lose, asking whether changing strategy might lose worse is the wrong question. The right question is, "does the alternate strategy have a higher chance of winning than the initial strategy, which is now behind?" That's why I described this concept initially as going for a high-risk, high-reward strategy. Yes, it will fail often, and when it does you will probably lose worse than you might have otherwise. But, if you're behind and there's almost no chance of catching up with your original strategy, then you want to take the small chance of salvaging a win from the jaws of defeat. It's the same logic behind player 4 going for a higher variance strategy by default in a multiplayer game; since their initial odds of winning are less than 25% (and sometimes substantially so), even a strategy that only works 25% of the time is an improvement in their odds.

[WanderingWinder]

I think you misinterpreted me. When I say "it could be worse", I don't mean you're losing worse, I mean you're more likely to lose.

[philosophyguy]

I think you misinterpreted me. When I say "it could be worse", I don't mean you're losing worse, I mean you're more likely to lose.

If that's what you meant, then that's obviously true. You don't go for a 20%-chance strategy unless your current strategy is worse than 20% odds. That's why you don't jump ship unless you're really far behind. The question I was asking is--what standard do top players use for judging when to abandon ship, since it's probably not possible to calculate the exact odds on the fly?

[tko]

Sometimes the game forces you into making a strategy change and you need to dig out a lucky alternative strategy.  For example, in this game, I Swindlered 1 of my opponent's Minions into a Dutchy on Turn 8 and bought my 5th Minion.  My opponent bought the remaining Minion, and doubly returned the favor in later turns by Swindlering 2 of my Minions into Dutchies.  In a dramatic shift, I go from being ahead in the Minion race 5-4 to falling behind 3-4.  I can no longer rely on Minion as my source of coin, and my opponent has purchasing power for Provinces and I can't dream of attaining that.  So I start loading up on Monuments and Hamlets, use Minion to cycle into my Monuments, and buy out the Dutchies (I already had 2 from Swindler) and 3-pile to end the game.  This game gave me an obvious cue that I need to go for luck.  Dominion is such a short game with such few turns, that unlike Magic the Gathering, it's very difficult to come back from being behind.  Then again, you can always play another.

[Superdad]

I think the idea of the original post has more to do with when you shoot for the stars (luck), not so much when you change gears - which is more of a decision based on re-evaluation of the gamestate.

Something like "I realised I was losing the tournament race, so I went for a duke/duchy ending" isn't so much what he's looking for (my interpretation of his post anyways). That's why my example had to do with planning your final few turns based on a certain combination of the remaining cards in your deck, because only that combination (or possibly, permutation) wins the game - and formulating your last buys based on it.

I guess the main thing to consider here is that there aren't *that* many high risk, high reward strategies in dominion. For the mostpart, the game is fairly balanced. Really, treasure map is the only true example of a super high-risk, high reward strategy that I can think of. Other strategies such as going for gardens when you are 0-3 in provinces (for another example) - that isn't really an example of going for luck, it's just changing gears.

Going for luck is more along the lines of picking up a coppersmith on turn 4 in a tournament game, and hoping your stars align to nab a quick province (if for example he hit a baron'd turn 3 province). In an example like this, if there is no real competitive strategy alternative than going for provinces, if he was lucky enough to get a turn 3 province, then your only recourse may be to Roll-the-dice and buy that coppersmith hoping to get a turn 5 province to compete.

Another example could be picking up a Kings Court when you don't have that many actions, but a kings courted montebank could pull you back in the game if the stars lined up. I'm not sure if that qualifies, since a kings court purchase with $7 is generally a good thing* (TM) regardless.

I think those are more relevant examples.

As far as when to go for luck... I think him sea-hagging your sea hag twice is unfortunate, but shouldn't result in a drastic swing in your strategy. Him Sabotaging/Theif/Pirate ship'ing your gold, swindler'ing your Montebank into a duchy... all those examples probably shouldn't trigger an overreaction, as they are are surmountable. You are likely doing yourself in by changing gears too quickly on something like that.  Now, if he does it 3+ times in a row, then maybe it's time to think about changing it up.

Whether you go for a new strategy (changing gears), or you shoot for the stars (picking up 2x treasure maps and crossing your fingers), whatever you do, it should be somewhat based on logical decision making. Try to quantify the probabilities.. the risk vs reward of the new strategy. I'd say that a triple swindler on $5's (turned into duchies) would be enough to make me consider it (for swindler especailly, possibly even a double swindler hitting $5 would be enough.. that's some rough beats right there).

[WanderingWinder]

Sometimes the game forces you into making a strategy change and you need to dig out a lucky alternative strategy.  For example, in this game, I Swindlered 1 of my opponent's Minions into a Dutchy on Turn 8 and bought my 5th Minion.  My opponent bought the remaining Minion, and doubly returned the favor in later turns by Swindlering 2 of my Minions into Dutchies.  In a dramatic shift, I go from being ahead in the Minion race 5-4 to falling behind 3-4.  I can no longer rely on Minion as my source of coin, and my opponent has purchasing power for Provinces and I can't dream of attaining that.  So I start loading up on Monuments and Hamlets, use Minion to cycle into my Monuments, and buy out the Dutchies (I already had 2 from Swindler) and 3-pile to end the game.  This game gave me an obvious cue that I need to go for luck.  Dominion is such a short game with such few turns, that unlike Magic the Gathering, it's very difficult to come back from being behind.  Then again, you can always play another.

I'm not sure that this is going for luck so much as switching to a better, though less obvious, strategy(!)

[philosophyguy]

I think the idea of the original post has more to do with when you shoot for the stars (luck), not so much when you change gears - which is more of a decision based on re-evaluation of the gamestate.

Something like "I realised I was losing the tournament race, so I went for a duke/duchy ending" isn't so much what he's looking for (my interpretation of his post anyways).

As the OP, I can say with confidence that your interpretation is not what I intended. It's certainly an interesting question (especially in multi-player games), but I was definitely thinking about "I've lost X race" or "my chosen strategy just blew up." Hence my initial examples of getting a Sea Hag flipped in turns 3-4 and losing a Minion race.

[Superdad]

But what is the next part?

I lost the race so I:

1) Changed strategies
or
2) Went for or started planning around luck (i.e. a high risk, high-reward strategy)?


If what you are looking for is #1, then the answer is pretty simple. You change gears as early as you feasibly can - provided your assumption of a destined-to-lose game is correct. The longer you wait, the more screwed you will become.

Good players can switch gears early. Bad players will switch gears too late and they'll actually decrease their odds of winning - both by a poor choice of when to switch gears, and likely also what to switch into.

Him sea hagging your sea hag twice (or any similarly tempo-negative impactful play) should cause an immediate re-evaluation of the gamestate, and a new best-strategy going forward should be formulated. This reaction should be immediate. If he swindlers two montebanks into two duchies in the first 6 turns of the game, I'm certainly changing gears to a quick 3-pile ending, or some form of remodel/upgrade/expand strategy. I would probably even change gears on the first swindle of my $5 into a duchy.

Him sea hagging me twice and hitting my sea hag.... that would cause me to re-evaluate my position in the game, but the likely reaction wouldn't be too drastic. For the most-part, my initial strategy is still likely the most probable for victory.

These reactions all have to come immediately, as the first 5-6 turns of the game your deck should be the most fluid, slowly getting more rigid (in terms of switching gears) each turn that passes.

/edit to further attempt to answer your question more completely... when is it too late to switch strategies? That will obviously depend on three things:

1) The secondary strategies available
2) The ability of your deck to mould into the new strategy.
3) The likelyhood of his deck stalling-out if you switch VP strategies.

For example, if you already have a warehouse and a horse traders, switching into duke/duchy isn't so bad (lets say you were going for minions). However, if you have 3 terminals and no villages are available, switching into a goons strategy just isn't going to work. That switch had to happen much earlier (if at all).

For point #3, lets say he's some form of Chapel deck that suffers greatly from greening up. If he will likely beat you to 4 provinces, and 3-pile ending isn't an issue, he will slow down a ton as he greens. If you can hit up some form of alternate Vp strat such as Vineyards, then you should go for it, if you are already down 3-0 (or even 2-0). You likely aren't winning a province race, and he won't be able to get 8 provinces, as he is succeptible to greening, so you'll have time to switch.

So the decision of when to switch gears is basically:

a) Do it as soon as you can - ideally. This can be plus/minus a few turns based on how easily your deck transforms into the new strategy
b) Only do it if you truly think you are behind. You will suffer some tempo loss for switching gears, and your new strategy has to have the potential to come back.

The times this is going to be the most successful is if your opponent is up say 3-0 in provinces and you were going to go for provinces. It's a lot harder for him to pick up all 8 provinces and end the game. If you can switch into some form of alternate VP game that is stronger the longer the game lasts (goons, vineyards/fairgrounds, gardens), and he will likely stall out if he is forced to get all 8 provinces, then you can probably go for broke and switch gears. But do so as early as you can.

[philosophyguy]

Excellent, Superdad; that's exactly the kind of explanation I was looking for.