I'm responding here to everything* in an attempt to get the discussion going again
I think the main thing about lowering piles is to think before you buy. One player once mentioned that this is a game about emptying piles whether provinces or three-piles. So, always remember that buying cards leads the game one-step closer to coming to an end.
Thinking before you buy sounds like great advice. Not just when lowering piles.
First thing that comes to mind is that situation:
You might want to do that when you are behind and your only hope is that your opponent duds on the next turn so you could potentially steal a win.
I think this is an important point. It's certainly one of the situations where I like to lower piles. But maybe you can go into more details? Are you doing this every time you're behind or are there more side conditions?
I like to do this thing when I'm in the lead, and I'm sure this is one of the errors you have in mind, which is lowering piles just to make the game end sooner even if it causes some slight decrease in win %.
And I was going to write more about how I agree with you on general principle, but I have seen cases where you've made comments along these lines and I didn't think it applied in that case. Which is to say, deciding what "being in the lead" means exactly and making estimations for likelihood of dud hands over a number of future turns isn't always obvious. But I don't have time now!
I hope you will find more time because you will certainly get it from me
I think this is really only something you're going to do if a) piles are lowering anyway, or at least one pile has already emptied and b) it looks like Provinces are going to be difficult to empty. This is most often the case in curser/looter games, but can also happen in engine games with a contested part. In these cases, you really have to be mindful of how much is left in each pile, and what your current score is. If you're losing, you want to keep the piles up to make it harder for your opponent to end the game in their favor.
Of course this is mainly about pile emptying. For pile lowering, I think the main card to look out for is Goons. With Goons, I'm perfectly happy grabbing a bunch of Pearl Divers, because they're basically cheaper Great Halls (or better, depending on how many Goons I have in play).
I dunno. Really, the strategy for pile lowering is the same as for pile emptying - you're looking for the same conditions, you're just starting earlier. In order for it work without a contested pile, you need some sort of +Buy or gainer.
Sorry, no, I'm not talking about Goons. I specifically said that lowering the pile wasn't going to benefit you. With goons in play you buy those cards because you want the points. I'm looking for scenario's where you want the piles to be lower.
Buy cards so that you will be able to win with empty piles on the following turn but your opponent will not. This threat does have to be backed up with enough vp to force a win though.
A similar situation might be to squeeze down the supply piles so that if the opponent buys their ideal cards they will offer you a 3 pile ending next turn. If you can lock an opponent out of buying a card like city they might be short of +buys or +actions for the rest of the game. This sort of squeeze can sometimes lock out the mid cost vp cards as well.
I think this points towards another major category of strategically lowering piles. Any chance you can formalize the conditions? Hmm maybe you already have.
I don't think Stef is talking about buying piles to win on your next turn though. He mentions lowering piles for the sake or lowering piles.
Oh yes winning on your next turn is fine. Winning on this turn is what I tried to ban with my talk about
not emptying pilesIn that case, let's talk about building vs closing. The first thing to look at will be how many times you will use your new cards if you decide to keep building. If there's only a 50-50 chance of playing them at all then building doesn't look so good. On the other hand, building may help you keep control over the important piles, typically the province pile, so that it doesn't actually matter whether the game closes sooner or later.
You're putting the whole discussion in it's proper context and I think that was certainly missing so far. In my estimation
lowering piles is a bit of an edge case. {buying points, building on, lowering piles}. Lowering piles is by far the least frequent one, I would estimate applies in <5% of the games.
So here's an anecdote about a RL game between 2 not-very-good players.
While I love anecdotes, it's hard to respond without knowing the complete board. You're suggesting your opponent should have emptied some other pile, which is possibly the right plan but not necessarily.
This definitely needs an article. More generally, I'd love to read a general article about end-game control, which is a (relatively) poorly understood aspect of advanced Dominion strategy. (I wrote an article for new players along these lines maybe a year ago, but I'd like to see an article for intermediate/advanced players.)
Some trivial observations:
-you shouldn't do it if you're behind
-you shouldn't do it if it your opponent can possibly buy enough points to overcome your lead and complete your 3-pile attempt
Some less trivial observations:
-most of the time when you do it, it doesn't matter, because you only do it when neither of the "trivial observations" apply, and "games where you're ahead and your opponent's deck can't buy a bunch of cards" is a pretty similar set to "games you're going to win anyway."
-the most important time to do it when it really does matter, is if your opponent threatens to gain an insurmountable lead, but doesn't quite have the engine to do it yet. This happens sometimes when you play a money-type variant against an engine-type variant. In this situation, you're "ahead" on points but "behind" on deck-tempo, another concept I'd like to see an advanced strategy article about.
The funny situation is that I agree with your less trivial observations and don't agree with your trivial ones. They nicely sum up the misconception I'm hoping to address.
The last one is an important point though, probably the most important one to make about lowering piles.
In this situation, you're "ahead" on points but "behind" on deck-tempo, another concept I'd like to see an advanced strategy article about.
In my mind at least, the deck quality (or chances of winning) is what I'm referring to by default when I talk about leads/behind/ahead. I usually will specify points if I'm talking about VP. I don't know if this is an important point of view to have, but it seems to me that lots of players still overvalue having their point counter be higher than their opponents at any stage of the game that isn't the end.
I think we're certainly going to need more than 1 metric to measure the gamestate when addressing this concept. Probably also more than 2?
I think it's easy for not-expert players to lower piles too early, so I try and be really conservative about it - I lower piles when I'm winning by more than one "big turn" worth of points. Lots of times when I miss three pile opportunities this way though.
I think that's a fine strategy at most levels of play. I often see players "going for the three pile" where just buying points would be much better (safer).
Here you're specifically ask about lowering piles in situations that don't lead to a forced win this turn or the next. I imagine you'd have to calculate the odds that this will lead to a 3-pile ending before your opponent overtakes you or you overtaking your opponent and ending the game on the same turn during some future turn.
I don't know how much you can say about this topic besides "depends on the game state". What I can recommend is to carefully consider the maximum number of gains your opponents can have on their turn before deciding to try lowering piles. If their gaining power matches yours, lowering piles can totally backfire if you get a dud hand relative to your opponent. I think it's better to lower piles when you have higher gaining power than your opponent so that you can lower piles only to the extent that it benefits you exclusively.
Edit: I'll add that it's better to lower piles that your opponents need for their strategy, because then you can count on your opponents to help you empty the piles as they build their engine. Simple example is lowering the Village pile when your opponent's strategy needs them, even if your strategy can make do with only a few of them.
Maximum number of gains is surely important somehow. Can we formalize the conditions around this?
I actually recommend you don't call it "lowering piles" because I think this term is misleading, because it's not clear if you're specifically referring to the end-game tactic of (eventually) emptying piles, or if you're just talking about lowering piles in general. Here are some examples of why I will want to lower/empty a pile, but I don't think they're what you're referring to by "lowering piles", right?
- you want as many of a pile as possible (Minion, Herald, etc.)
- upgrading Cities
- denying your opponent a pile
- getting access to a different Knight/Ruin
No I was not referring to any of these. I don't think it was misleading either, as I specified
for the mere sake of getting closer to a 3-pileBut feel free to suggest a better term.
To make good decisions about lowering piles I think you need to be able to track decks well too. Knowing what your deck and your opponent's deck are capable of is pretty much the most important information you can use to decide which piles to lower, and by how much, and when. At least it seems that way to me.
I'm now thinking there are a couple of fundamentally different scenario's where lowering piles is good. Exactly knowing both the decks capabilities is very important in some of them but not all.
Another type of pile lowering that doesn't gain you points is milling a Province. Rebuild is an obvious example but Remodel and Expand do it too. I'm not talking about ending the game this way but early/mid game. I know I have done this and it has hurt me and helped me win games in the past.
Milling a province is an excellent example of lowering piles yes. As is buying a curse with Watchtower in hand on a board with no cursers.
...
You make or try to make so many different points that it's hard to respond.
About fewer dice rolls: I agree that sometimes you want fewer dice rolls, but there are also conditions where you want more dice rolls (if you die is better). Can we formalize these conditions?
About VP gaining speed and its derivative: I really don't like using formulas for the mere sake of using formulas. Either you introduce and use them or just use words. I think you have exactly one valid point (If your opponent is going to outscore you in the long run better end it soon) that doesn't need any formulas to make. And those formulas are probably wrong, because the vast majority of "exponential decks" aren't scoring points at all for a long time - they assemble something else.
The asymmetries you are talking about are certainly interesting. I think they mostly resolve around a trade-of between increasing your decks power and increasing the maximum number of gains. Sometimes you'll sacrifice a bit of power to enable some tricks (Stonemason indeed, but also Develop is great in this spot). In these circumstances
Lowering Piles is probably something you do in addition to increasing your maximum number of gains to make the threat effective.