Dominion Strategy Forum
Dominion => Dominion General Discussion => Topic started by: kameraDSM on June 18, 2011, 10:38:45 am
-
Hey forum, I'm relatively new to Dominion and have loved every game on Isotropic and with friends - many of them humbling losses.
My question to those experienced players reading this: What is considered an efficient/fast win turn-wise? Naturally this varies widely based on the type of game (Curse, Colony, trash heavy, etc) so perhaps you might break it down that way.
Conversely, at which turn number is it clear that the players are widely inexperienced or inefficient?
And a bonus question: What's the longest game you've experienced?
Thanks to all!
-
Here is the turn count for my last 10 games (not in order): 15, 19, 19, 20, 20, 21, 21, 23, 25, 29
Here are are Dstern's (current #1 player on isotropic): 15, 17, 18, 20, 22, 22, 22, 30, 31, 34
This is of course highly anecdotal, and I'd love to see some more thorough stats on this from councilroom -- rrenaud? Maybe an 'average game length' stat on player pages? Maybe average game length as a column in the card effect table? That would actually be super useful, I could make a list of !<every card that takes more than 30 turns> to use when I'm in a hurry :D
Being able to see a histogram of a player's game lengths would be cool too, or at least min/max/mode in addition to mean.
-
Interesting insights! It would be helpful to know what type of games yield average turn results, both for time-saving reasons and as a personal measuring stick of game management.
-
back when i was more of a newbie, i'd had games up to 50 turns
-
Without attacks, you should be able to always get to 4 Provinces in ~17 turns because that's the baseline for just buying Treasure. If there's Smithy or Envoy on the board you can get to 4 Provinces in ~14 turns with just Treasure and a single copy of Smithy or Envoy.
-
You can also see the average length of games when you (and Isotropic in general) buy/gain any particular card on your Popular Buys page. (http://councilroom.com/popular_buys?player=kameraDSM)
-
Thanks for the info, theory, and for setting up the forum.
-
Well, sometimes the game will last longer, even if I could end it earlier. Just because I am trying to get this game safely won and not getting wrecked by extreme lucky draws from my opponents.
-
Well, sometimes the game will last longer, even if I could end it earlier. Just because I am trying to get this game safely won and not getting wrecked by extreme lucky draws from my opponents.
I'm curious exactly what you mean here. Are you just referring to buying duchies until your opponent doesn't answer (and breaks the PPR)? I think this is one the weakest aspects of my game; closing in a 'safely' in mirrors. Especially when my deck is something that's not resistant to bloat (duchies).
-
I think DStern only plays colony games, which will skew his numbers up generally
-
Well, sometimes the game will last longer, even if I could end it earlier. Just because I am trying to get this game safely won and not getting wrecked by extreme lucky draws from my opponents.
I'm curious exactly what you mean here. Are you just referring to buying duchies until your opponent doesn't answer (and breaks the PPR)? I think this is one the weakest aspects of my game; closing in a 'safely' in mirrors. Especially when my deck is something that's not resistant to bloat (duchies).
I assume what he means is, for instance, you have a 1province lead in a really bogged down game and the third pile has 3 left. You have only one buy and could spend three turns buying them out, or you can grab another Province or two for insurance before trying to empty the pile.
-
What theory said, +not buying any provinces when it isnt your strategy (gardens, goons+village,duchy-duke, vineyard, etc) even if you could. This would just give the opponent the chance on maybe getting lucky and grab the last provinces, where your own strategy has even more potential for VP points or you are behind in points.
Its not really easy to resist a Province and I have made this mistake many times too...
Even yesterday i broke the PPR for colonies. It was pure retardedness?! and lost me the game :-)
Endgame play can be really unnerving when you see your mistake after the game
-
I think DStern only plays colony games, which will skew his numbers up generally
Heh, this is true!
You can also see the average length of games when you (and Isotropic in general) buy/gain any particular card on your Popular Buys page. (http://councilroom.com/popular_buys?player=kameraDSM)
Oh, cool, thanks! My numbers are pretty much all basically 20, even in Colony games. Bizarrely, the biggest difference between colony-and non-colony games I have is for Young Witch, which is 20.1 turns in Province games, and 16.9 in colony games. I thought this was going to be due to a tiny sample size, but I have played 15 colony games with Young Witch.. I guess that's still not very many, but could it somehow be a real effect? It's certainly not true for isotropic overall...
Here's something interesting: exactly 3 cards have a colony-game average turn count less than 22. One of them is Chapel, which should come as no surprise. The other two are maybe not what you would expect, though they are related. Take a guess before you peek:
Card-drawing durations: Wharf and Caravan. All three of them are in the 21.7s, but Wharf is actually the fastest, at 21.7 even. Caravan is 21.72, and Chapel is 21.74. Caravan and Wharf switch spots in Province games, where Caravan makes it all the way down to 19.94, the only card in Dominion to break the 20-turn barrier.
-
Does anyone know exactly how councilroom.com calculates those stats? Because according to my player page, every single card averages a shorter game length with colonies than in general.
http://councilroom.com/popular_buys?player=ycz6
But in almost every case, the (mean + standard deviation) for Turns (C) is higher than for Turns. Which makes me wonder if that statistic is calculated the same way that the Isotropic leaderboard is? That still doesn't make any sense... and is also not how means work.
Edit: Actually, scratch that. I didn't look very hard at the data. The cards with shorter average turns have lower scores for Colony games than otherwise, while the cards with higher average game lengths have higher scores for Colony games, though still not by much.
-
I'm still confused by what you mean. Your page seems to make sense: with Mountebank, the average Province game takes 24.49 ± 1.76 turns and 25.17 ± 3.60 turns in Colony games.
-
Erm, try looking at more than one card maybe?
Baron: 19.60 ± 1.64, 13.00 ± 7.90 with Colonies
Ambassador: 21.23 ± 1.94, 15.33 ± 9.73 with Colonies
Menagerie: 19.91 ± 3.41, 12.83 ± 8.14 with Colonies
Not to mention some of these stats are blatantly wrong... only one of my 4 Menagerie + Colony games ended in fewer than 15 turns, and that one was a resignation. These stats can't actually be mean ± standard deviation.
-
There is a bug in the stats computing code. The code that calculates standard deviations assumes its being used for win rates, and adds a smoothing factor (it pretends you went 1-1 with any card in 2 2p games, even if it hasn't seen any data) which helps a lot with win rate stats where the averages should be around 1, but then it pulls the game length stats towards 1.
-
I fixed the crazy game length stats by https://github.com/rrenaud/dominionstats/commit/2e2e79dab4d29a58d869474a9872830962b81925
Hopefully the fix doesn't break other things.
-
Looks fixed to me! No more obvious errors, at least; there are still a few instances of averaging longer games with Colonies, but those are likely just due to random variation. Thanks :)
-
Back to the OP...to answer your question, one would have to know what cards are in play, and also the nature of the players.
Colonies will typically add a couple turns. Attacks that issue Curses or force players to discard will likewise lengthen the game. If Goons is in play, the game may go longer because players aren't as quick to purchase Victory cards, and also because they are clogging their decks with Coppers. If someone is playing a Gardens strategy, he will try to shorten the game. Draw engines will speed the game up because the good cards are going into your hand sooner. Trashing cards, especially Chapel, will make games significantly shorter.
The other variable is your opponent. Some players prefer winning it all on one turn using long action chains and multiple buys, but that usually takes a few turns to set up. Others go for a quick-and-dirty strategy. Which approach will win is a function of the cards that are in place, and the luck of the draw.