Dominion Strategy Forum

Please login or register.

Login with username, password and session length
Pages: 1 [2] 3 4 ... 8  All

Author Topic: [Discussion] DominionStrategy Qualifier for 2012 US National Championships  (Read 25412 times)

0 Members and 1 Guest are viewing this topic.

theory

  • Administrator
  • *****
  • Offline Offline
  • Posts: 3594
  • Respect: +6034
    • View Profile
    • Dominion Strategy
Re: Re: DominionStrategy Qualifying Day Results
« Reply #25 on: June 24, 2012, 07:51:00 pm »
0

Yeah -- winning three player pods is obviously easier than winning four player pods, but it is unfair to those randomly drawn into a three-player pod if it is impossible for them to score as high as someone drawn in a four player pod.
Logged

zxcvbn2

  • Golem
  • ****
  • Offline Offline
  • Posts: 193
  • Respect: +87
    • View Profile
Re: Re: DominionStrategy Qualifying Day Results
« Reply #26 on: June 24, 2012, 07:51:23 pm »
0

zxcvbn's proposal is still unbalanced, it just gives the three player pods the shaft now. Winning three of 4 games should be enough for the three player pod winners to advance. I think the point values need adjustment, but its hard to find the right spot where all the outcomes seem reasonable.

You have a 1 in 3 shot at winning a 3 player game. you have a 1 in 4 shot of winning a 4 player game. I fail to see how it's unfair to award 3 points for winning a 3 player game, and 4 for winning a 4 player game.
Logged
One Day Cup II Champion: qmech III: Rabid IV: Qvist

zxcvbn2

  • Golem
  • ****
  • Offline Offline
  • Posts: 193
  • Respect: +87
    • View Profile
Re: Re: DominionStrategy Qualifying Day Results
« Reply #27 on: June 24, 2012, 07:53:01 pm »
0

Yeah -- winning three player pods is obviously easier than winning four player pods, but it is unfair to those randomly drawn into a three-player pod if it is impossible for them to score as high as someone drawn in a four player pod.

But it's that much more difficult to score that much higher in a 4 player game. You have that much tougher of a time winning 3 games in a four player match than in a 3 player match. You're so much more likely to win 3 games in a 3 player match than in a 4 player match.

Obviously, this isn't my call, and I understand your perspective. There is just SO much difference between 3 and 4 players.
« Last Edit: June 24, 2012, 07:55:39 pm by zxcvbn2 »
Logged
One Day Cup II Champion: qmech III: Rabid IV: Qvist

theory

  • Administrator
  • *****
  • Offline Offline
  • Posts: 3594
  • Respect: +6034
    • View Profile
    • Dominion Strategy
Re: Re: DominionStrategy Qualifying Day Results
« Reply #28 on: June 24, 2012, 07:57:39 pm »
0

I think we have two separate debates going on here, so let's clarify.

4p vs 3p

Obviously we'd love to minimize 3p games.  But we'll have them, and either you get screwed if you are in a 3p game or you get a benefit.  I think the current system offers a small benefit, with the caveat that you basically lose all ties with anyone who got the same number of wins in 4p games.

In other words, if I had to choose between a 3p player who went 1-1-2 and a 4p player who went 1-2-2-2, I will go with the 3p player, and if the 4p player went 1-1-2-4, I'll go with the 4p player.

Points vs wins

I really don't think this makes any difference at all.  Most point scoring systems are functionally equivalent to wins-matter-most, with second places only being a tiebreaker.  If you can point out an example of a situation where a fair points system would differ from a wins system and lead to a clear inequity, then I may change my mind.
Logged

rrenaud

  • Administrator
  • *****
  • Offline Offline
  • Posts: 987
  • Uncivilized Barbarian of Statistics
  • Respect: +1177
    • View Profile
    • CouncilRoom
Re: Re: DominionStrategy Qualifying Day Results
« Reply #29 on: June 24, 2012, 08:09:35 pm »
0

The nice thing about winner take all is that all opps have same objective.  If one player is playing for 2nd or 3rd, they can dramatically change the pace.  It introduces more tournament meta gaming. 

Having walk on alts to fill up 3p games might be nice though.
Logged

Mic Qsenoch

  • 2015 DS Champion
  • *
  • Offline Offline
  • Posts: 1673
  • Respect: +4275
    • View Profile
Re: Re: DominionStrategy Qualifying Day Results
« Reply #30 on: June 24, 2012, 08:10:06 pm »
0

You have a 1 in 3 shot at winning a 3 player game. you have a 1 in 4 shot of winning a 4 player game. I fail to see how it's unfair to award 3 points for winning a 3 player game, and 4 for winning a 4 player game.
Those aren't the only points awarded in your system though. A second place (4 player) finish is worth twice as many points as a second place (3p) finish in one of your schemes. I'm not sure there's a way to balance the points between the two types of pods.

Also theory, I was hoping you could clarify how ties within a game are being accounted for. Are ties for first worth as much as a standalone win, and do ties at say 3-4 count as 3rd place finishes or something in between 3rd and 4th? Mostly I'm wondering how this is considered for final tiebreak situations.
Logged

Kirian

  • Adventurer
  • ******
  • Offline Offline
  • Posts: 7092
  • Shuffle iT Username: Kirian
  • An Unbalanced Equation
  • Respect: +9371
    • View Profile
Re: Re: DominionStrategy Qualifying Day Results
« Reply #31 on: June 24, 2012, 08:11:38 pm »
+1

Points vs wins

I really don't think this makes any difference at all.  Most point scoring systems are functionally equivalent to wins-matter-most, with second places only being a tiebreaker.  If you can point out an example of a situation where a fair points system would differ from a wins system and lead to a clear inequity, then I may change my mind.

Player 1:  Takes 1st twice and 4th twice.  5 + 5 + 1 + 1 = 12.
Player 2:  Takes 1st once and 2nd thrice:  5 + 3 + 3 + 3 = 14.
Player 3:  Takes 1st once, 2nd twice, and 3rd once:  5 + 3 + 3 + 2 = 13

Obviously player 2 is better than player 3 by any measure.  Tougher to say if player 1 is truly better than the other, though.  And that's ignoring seating troubles.
Logged
Kirian's Law of f.DS jokes:  Any sufficiently unexplained joke is indistinguishable from serious conversation.

WanderingWinder

  • Adventurer
  • ******
  • Offline Offline
  • Posts: 5275
  • ...doesn't really matter to me
  • Respect: +4368
    • View Profile
    • WanderingWinder YouTube Page
Re: Re: DominionStrategy Qualifying Day Results
« Reply #32 on: June 24, 2012, 08:12:29 pm »
+2

The nice thing about winner take all is that all opps have same objective.  If one player is playing for 2nd or 3rd, they can dramatically change the pace.  It introduces more tournament meta gaming. 

Having walk on alts to fill up 3p games might be nice though.
Of course, I disagree with this assessment entirely, and think that there's more problems of this ilk with a winner-take-all system. But we've been through this multiple multiple times now.

rrenaud

  • Administrator
  • *****
  • Offline Offline
  • Posts: 987
  • Uncivilized Barbarian of Statistics
  • Respect: +1177
    • View Profile
    • CouncilRoom
Re: Re: DominionStrategy Qualifying Day Results
« Reply #33 on: June 24, 2012, 08:14:50 pm »
0

This is a different claim.  It's not about settling for 2nd in a particular game.

It's about gunning for 2nd in the 3rd or 4th game of the tournament from the start of the game. 
Logged

WanderingWinder

  • Adventurer
  • ******
  • Offline Offline
  • Posts: 5275
  • ...doesn't really matter to me
  • Respect: +4368
    • View Profile
    • WanderingWinder YouTube Page
Re: Re: DominionStrategy Qualifying Day Results
« Reply #34 on: June 24, 2012, 08:19:07 pm »
0

I don't see any meaningful way to AIM for 2nd from the beginning of a game, rather than playing what you think to be the best strategy anyway.
Edit: Moreover, I don't see the problem with that.
« Last Edit: June 24, 2012, 08:24:44 pm by WanderingWinder »
Logged

zxcvbn2

  • Golem
  • ****
  • Offline Offline
  • Posts: 193
  • Respect: +87
    • View Profile
Re: Re: DominionStrategy Qualifying Day Results
« Reply #35 on: June 24, 2012, 08:33:43 pm »
0

In a 3 player game you are twice as likely to win at least 3 games as in a 4 player game. (1/9 as opposed to 13/256) Maybe it's just me, but I see this as more than a small benefit. Yes, winning 4 games has is worth 4/3 as much in my setup, but it's also also over 3 times more likely. (Obviously, these are all assuming equal ability.)

Perhaps my scoring method could have been tweaked to award 2nd player in 3p games more points (indeed this was my intention with the second method). I still think it was fairer than what we have, though I really don't want to continue arguing after this post, because I do believe we should keep the format we started with for the sake of continuity.
Logged
One Day Cup II Champion: qmech III: Rabid IV: Qvist

popsofctown

  • Adventurer
  • ******
  • Offline Offline
  • Posts: 5359
  • Respect: +2755
    • View Profile
Re: Re: DominionStrategy Qualifying Day Results
« Reply #36 on: June 24, 2012, 08:38:13 pm »
0

I don't see any meaningful way to AIM for 2nd from the beginning of a game, rather than playing what you think to be the best strategy anyway.
Edit: Moreover, I don't see the problem with that.
If you're aiming for second you're less likely to go for "lottery ticket" strategies like turn 2 and 3 Treasure Maps.  You can just play a stable game and take second.
Logged
Also you probably are an expert if you buy two bureaucrats early.

timchen

  • Minion
  • *****
  • Offline Offline
  • Posts: 704
  • Shuffle iT Username: allfail
  • Respect: +233
    • View Profile
Re: Re: DominionStrategy Qualifying Day Results
« Reply #37 on: June 24, 2012, 08:50:03 pm »
0

Is there any seeding for how to separate players into pods? If it is random it sounds like a larger problem for me. That is, if someone wins 4 out of 4, I imagine it tells more that his opponents are weak rather than he is super strong. This is the strongest problem I see for a winner-take-all format. Basically 2 equally strong players will destroy each other with similar luck.

I therefore think a point system is better. Like match points. Convert how many opponents you have beaten into percentages. So 4p: 100% 67% 33% 0%, 3p:100% 50% 0%. I have to say I feel this points reflects skill better than the sole winner.
« Last Edit: June 24, 2012, 09:13:24 pm by timchen »
Logged

rrenaud

  • Administrator
  • *****
  • Offline Offline
  • Posts: 987
  • Uncivilized Barbarian of Statistics
  • Respect: +1177
    • View Profile
    • CouncilRoom
Re: Re: DominionStrategy Qualifying Day Results
« Reply #38 on: June 24, 2012, 08:53:15 pm »
0

We seeded by iso rank, so the top N/4 players didn't have any other the other top N/4 in their pods.
Logged

timchen

  • Minion
  • *****
  • Offline Offline
  • Posts: 704
  • Shuffle iT Username: allfail
  • Respect: +233
    • View Profile
Re: Re: DominionStrategy Qualifying Day Results
« Reply #39 on: June 24, 2012, 09:12:32 pm »
0

but judging from the results, the (mainly) 2p ranking is probably not a good way to seed...
Logged

zxcvbn2

  • Golem
  • ****
  • Offline Offline
  • Posts: 193
  • Respect: +87
    • View Profile
Re: Re: DominionStrategy Qualifying Day Results
« Reply #40 on: June 24, 2012, 09:13:16 pm »
0

but judging from the results, the (mainly) 2p ranking is probably not a good way to seed...

This was a very small sample size.
Logged
One Day Cup II Champion: qmech III: Rabid IV: Qvist

zxcvbn2

  • Golem
  • ****
  • Offline Offline
  • Posts: 193
  • Respect: +87
    • View Profile
Re: Re: DominionStrategy Qualifying Day Results
« Reply #41 on: June 24, 2012, 09:20:58 pm »
0

Also, I'm just talking and don't want anything changed, but have just ONE more thing. I think, if we would have rotated opponents, the advantage of those playing 3p matches would have been spread out and we wouldn't have that much of a problem.

.

.


.

Okay, now I'm done talking about it. No, really.
Logged
One Day Cup II Champion: qmech III: Rabid IV: Qvist

metzgerism

  • Baron
  • ****
  • Offline Offline
  • Posts: 56
  • Respect: +15
    • View Profile
Re: Re: DominionStrategy Qualifying Day Results
« Reply #42 on: June 24, 2012, 09:21:53 pm »
0

theory: All I remember before this, in terms of what could be construed as a "decree," was this:

Quote from: theory
This is completely overblown, this points-vs-winners thing.

Let's be realistic.  With any of the points systems we're considering, (5-3-2-1), those with 4 wins will always beat those with 3 wins will always beat those with 2 wins.  So this is a completely unnecessary distinction at this point.

Although a lot of players today probably just played as best as they could, some might not have. Many of us agree that whatever system is used encourages playing towards that system, and not necessarily Dominion in general. Because there wasn't anything set in stone beforehand, you should award spots based on most wins, with collective non-win placements as tiebreaks.

Re: Seeding @ timchen:

If you were to seed based on 4p rankings, you're going to have a lot of participants just flat-out unseeded (and those might be the best players of all). Since 2p is the primary way we play, 2p is the only logical & available way to separate players.
Logged

metzgerism

  • Baron
  • ****
  • Offline Offline
  • Posts: 56
  • Respect: +15
    • View Profile
Re: Re: DominionStrategy Qualifying Day Results
« Reply #43 on: June 24, 2012, 09:22:28 pm »
0

Also, I'm just talking and don't want anything changed, but have just ONE more thing. I think, if we would have rotated opponents, the advantage of those playing 3p matches would have been spread out and we wouldn't have that much of a problem.

.

.


.

Okay, now I'm done talking about it. No, really.
I think you get to try again on Tuesday if you failed to qualify today. Is that correct? Is that also enough rotation for you?
Logged

timchen

  • Minion
  • *****
  • Offline Offline
  • Posts: 704
  • Shuffle iT Username: allfail
  • Respect: +233
    • View Profile
Re: Re: DominionStrategy Qualifying Day Results
« Reply #44 on: June 24, 2012, 09:25:07 pm »
0

Quote
If you were to seed based on 4p rankings, you're going to have a lot of participants just flat-out unseeded (and those might be the best players of all). Since 2p is the primary way we play, 2p is the only logical & available way to separate players.

I know. I am just saying, a match point system can alleviate this problem. :3 Also it is fairer w.r.t 3p and 4p games.
Logged

zxcvbn2

  • Golem
  • ****
  • Offline Offline
  • Posts: 193
  • Respect: +87
    • View Profile
Re: Re: DominionStrategy Qualifying Day Results
« Reply #45 on: June 24, 2012, 09:25:29 pm »
0

Also, I'm just talking and don't want anything changed, but have just ONE more thing. I think, if we would have rotated opponents, the advantage of those playing 3p matches would have been spread out and we wouldn't have that much of a problem.

.

.


.

Okay, now I'm done talking about it. No, really.
I think you get to try again on Tuesday if you failed to qualify today. Is that correct? Is that also enough rotation for you?

You misunderstand me. I don't think there was a problem with playing the same opponent four times. I think many people playing 3 4p games and 1 3p game is more fair than a few playing 4 3p games.
Logged
One Day Cup II Champion: qmech III: Rabid IV: Qvist

metzgerism

  • Baron
  • ****
  • Offline Offline
  • Posts: 56
  • Respect: +15
    • View Profile
Re: Re: DominionStrategy Qualifying Day Results
« Reply #46 on: June 24, 2012, 09:35:36 pm »
+1

I'm going to go out on a limb here and suggest again that multiplayer winner-take-all is the most arbitrary system that could have been used, that still involved actual gameplay.  And I'm actually not convinced that using a poll would have been more arbitrary, especially given the total of only four games used, all with the same four people.  Not a single qualifier demonstrated superiority of play over more than three people here.

The seat rotation method used was poor too, since we're bringing up flaws.  With two excellent players and two people who bought cards at random in a pod, it was possible for one of the excellent players to go before the other one three times.
Are you sure you're using "arbitrary" correctly?

I agree about the seat rotation being pretty harsh - there's 24 different seating orders that could be used, though, it's not like it's easy to figure out what's going to be fair. Here's what I'd suggest:

A-B-C-D
D-C-B-A
B-A-D-C
C-D-A-B

Everyone gets every seat, and you're only sitting before/after the same person in games 1 and 4.
Logged

metzgerism

  • Baron
  • ****
  • Offline Offline
  • Posts: 56
  • Respect: +15
    • View Profile
Re: Re: DominionStrategy Qualifying Day Results
« Reply #47 on: June 24, 2012, 09:37:03 pm »
0

Also, I'm just talking and don't want anything changed, but have just ONE more thing. I think, if we would have rotated opponents, the advantage of those playing 3p matches would have been spread out and we wouldn't have that much of a problem (This isn't a very serious proposal, btw).

.

.


.

Okay, now I'm done talking about it. No, really.
I think you get to try again on Tuesday if you failed to qualify today. Is that correct? Is that also enough rotation for you?

You misunderstand me. I don't think there was a problem with playing the same opponent four times. I think many people playing 3 4p games and 1 3p game is more fair than a few playing 4 3p games.
Oh, yes. I'm actually not delving into the 3p vs. 4p debate quite yet because I don't have a great solution for that. My recommendation would definitely be to have backup players.

Honestly, there's not a good objective way to do the 3p pods. I think the 3p results need to just count 2/3 as much, which means that they have to pull 3 wins to have a shot. Again...it's not ideal or fair, but it's not fair for someone any way it is right now.
« Last Edit: June 24, 2012, 09:48:47 pm by metzgerism »
Logged

Kirian

  • Adventurer
  • ******
  • Offline Offline
  • Posts: 7092
  • Shuffle iT Username: Kirian
  • An Unbalanced Equation
  • Respect: +9371
    • View Profile
Re: Re: DominionStrategy Qualifying Day Results
« Reply #48 on: June 24, 2012, 10:10:25 pm »
+1

I'm going to go out on a limb here and suggest again that multiplayer winner-take-all is the most arbitrary system that could have been used, that still involved actual gameplay.  And I'm actually not convinced that using a poll would have been more arbitrary, especially given the total of only four games used, all with the same four people.  Not a single qualifier demonstrated superiority of play over more than three people here.

The seat rotation method used was poor too, since we're bringing up flaws.  With two excellent players and two people who bought cards at random in a pod, it was possible for one of the excellent players to go before the other one three times.
Are you sure you're using "arbitrary" correctly?

See definition 4.

Quote
Here's what I'd suggest:

A-B-C-D
D-C-B-A
B-A-D-C
C-D-A-B

Everyone gets every seat, and you're only sitting before/after the same person in games 1 and 4.

And since the person you're directly before or after doesn't make much difference, that last part isn't an issue.  What you've posted is perfect, as every player goes before every other player twice.  Thanks for taking the time to actually do that.
Logged
Kirian's Law of f.DS jokes:  Any sufficiently unexplained joke is indistinguishable from serious conversation.

metzgerism

  • Baron
  • ****
  • Offline Offline
  • Posts: 56
  • Respect: +15
    • View Profile
Re: Re: DominionStrategy Qualifying Day Results
« Reply #49 on: June 24, 2012, 10:25:20 pm »
0

I'm going to go out on a limb here and suggest again that multiplayer winner-take-all is the most arbitrary system that could have been used, that still involved actual gameplay.  And I'm actually not convinced that using a poll would have been more arbitrary, especially given the total of only four games used, all with the same four people.  Not a single qualifier demonstrated superiority of play over more than three people here.

The seat rotation method used was poor too, since we're bringing up flaws.  With two excellent players and two people who bought cards at random in a pod, it was possible for one of the excellent players to go before the other one three times.
Are you sure you're using "arbitrary" correctly?

See definition 4.

Quote
Here's what I'd suggest:

A-B-C-D
D-C-B-A
B-A-D-C
C-D-A-B

Everyone gets every seat, and you're only sitting before/after the same person in games 1 and 4.

And since the person you're directly before or after doesn't make much difference, that last part isn't an issue.  What you've posted is perfect, as every player goes before every other player twice.  Thanks for taking the time to actually do that.
You're welcome! Made sense to me :) Although I disagree about the "who you're sitting before and after" part - Possession, Masquerade, and final turn dynamics care about those things.

---

I disagree strongly to your assessment that winner-take-all is arbitrary based on that definition of the term, and now I have an ace-in-the-hole reason:

From the Dominion rulebook, page 8: "Game End"
Quote
The player with the most victory points wins.
If the highest scores are tied at the end of the game, the tied player who has
had the fewest turns wins the game.
If the tied players have had the same number of turns, they rejoice in their shared victory.

Nowhere in the rulebook does it state anything about players who did not finish with the most points, only that they were not victorious. There's no talk about second-place points, third-place, etc. None.

The only distinction made is between players who won, and players who did not win. According to the actual Dominion rules, there's only one prize handed out: victory (shared or outright). Therefore, winner-take-all seems like the LEAST arbitrary system to employ, in my observation and opinion.

EDIT: I don't really want to open this can of worms again, but the use of "arbitrary" really bristled me, and I just remembered to use the rules as evidence. So I have.
« Last Edit: June 24, 2012, 10:28:06 pm by metzgerism »
Logged
Pages: 1 [2] 3 4 ... 8  All
 

Page created in 0.11 seconds with 21 queries.