Print

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

[theory]

We would like to host a 3p qualifying tournament for US Nationals on Isotropic.  It must conclude by next week, or at the latest, sometime in the first week of July because US Nationals will be in Chicago in mid-July.  The winner receives lodging and airfare to Nationals.

We want to solicit your ideas on how to organize this tournament.  Time is obviously not on our side.  How can we handle signups, matchups, and no-shows in a fair and equitable manner?

[Ozle]

Airfare?! Im assuming thats only within USA?

[theory]

Yes.  Surprisingly, "US National Championships" is only open to Americans ;-)

I have advocated for us to host an "international qualifier", but Rio Grande is unwilling (somewhat understandably).  We can only do an American qualifier.

[WanderingWinder]

First, you want to make sure everyone who enters can actually use the prize. With work schedules, and needing to get time off, you've got to get the tournament done soon enough that people who will need some notice for this will be able to get the time off. It would also be extremely helpful to know just when in mid-July the nationals are.

3p... well, you've got some logistical issues here. All the players are US, which will help anyway. But I think general consensus would be that you need more games, not fewer, to have really fair outcomes for 3p. Well, maybe there's not enough time for this, but more games is still going to be better than fewer.

Specifying particular matchups might actually be a huge problem here, as it's really going to put a hamper on how quickly you can do things. If you specify matchups, you get... one? Two? Sets of matches that you can reasonably expect? Very few, anyway.
OTOH, if you do a more open format, you have issues with fairness there, making sure everybody can get the same number of games in, etc. etc. Which is hard.
I'm not sure I see a great solution here, but it might help to know roughly how many people are going to be competing here. It's a lot different trying to cull out from 9 to 1, than it is from 100 down to 1.

You may need to do some heavy-duty single days. Which sucks for people unavailable those days, but them's the breaks, with so little time.

No-shows are probably just out. No hard feelings, but we don't really have time for something else.

Will continue to vomit ideas onto page as I have them.

[Ozle]

I can pretend to be american....get me a stetson and start saying Howdy Y'all!

And i think you are going to have to be tough on the formats, no trying to please everybody. Just pick something quick, set the dates and if people cant make it, tough

[ednever]

Some quick thoughts:

(1) Coordinating play times for 3-P is tough. Especially in a limited timeline. This suggests you set some times and people either play or they don't (like a real in-person tournament)

(2) Time zones are tough. Setting up a 1-day tourney that works in Europe is tough for the west coast of the US (and almost impossible for Asia). This suggests you may want a 'first round' split into two time brackets

(3) Single elimination is really random in 3P. Being the third player could take you out right off the start. This suggests you want multiple plays before any elimination (and in each elimination round)


From that, and given the time limitations, here's a suggestion:

(1) Two prelim rounds: One starts Sat morning GMT. The other starts Sat Morning PST
(2) Top 6 players in each prelim play again on Sunday
(3) For prelims set up multiple rounds where points determine moving to the next round. You can eliminate any number of people each round to make it work
(4) For the finals the next day, have the 6-players play twice in each spot. Most points win.

One more thought:
I played in a Magic tournament years and years ago with the following format:
- Find any opponent you haven't played yet and play a game
- When you finish a game, find a new opponent
- You get 1 point for a loss and 2 points for a win

Maybe not right here, but it encouraged people to go out and play a lot (and play fast) over the course of the day (In Magic is led to the smart people building really fast decks, but that's obviously not possible with Dominion).

It could be that you just do the first round that way as a clearing mechanism. 2 points for a win, 1 for 2nd place. Go play as many games as you can with new people for 3-5 hours. The top players move on to a second, more structured round? Top three in the structured round move to the finals the next day.

Lots wrong with this, but feel free to use it as a starting point for something you could get going fast...

Ed

EDIT: Obviously without Europe competing you don't need the EU timezone... My bad.

[Ozle]

5 qualifying days over the evenings, give people the chance to pick and choose, and then finals on the sunday, giving you more time to play.

I would suggest people put the qualifying days they can play on the sign p, and you assign them to a day as best you can to achieve roughly even numbers.

Top person from each mon-sat heat goes into 6 person finals on the sunday.
Obviously amount of heat games depends on how many sign up

[Axxle]

The Nationals conflicts with San Diego Comic-Con so I won't be participating.  I just wanted to wish everyone luck in the qualifier and in the Championship if you do end up going!

[Fabian]

After hearing that Swedish nationals wouldn't receive sponsorship to fly to Gencon, I'm feeling extra super jealous to have a chance to qualify (to anything) online Hopefully next year there will be something similar open to non-americans, it would be cool.

[rrenaud]

3p... well, you've got some logistical issues here. All the players are US, which will help anyway. But I think general consensus would be that you need more games, not fewer, to have really fair outcomes for 3p. Well, maybe there's not enough time for this, but more games is still going to be better than fewer.

The winner probably won't be the best player.  If we wanted the best American player, I'd happily take the top player from the iso leaderboard.

I don't think we should shoot for maximizing the chance of selecting the best player.  I think we should try to make it fun, get one of you lucky forum members a chance to get to the US nationals, and making it actually happen.

[WanderingWinder]

3p... well, you've got some logistical issues here. All the players are US, which will help anyway. But I think general consensus would be that you need more games, not fewer, to have really fair outcomes for 3p. Well, maybe there's not enough time for this, but more games is still going to be better than fewer.

The winner probably won't be the best player.  If we wanted the best American player, I'd happily take the top player from the iso leaderboard.

I don't think we should shoot for maximizing the chance of selecting the best player.  I think we should try to make it fun, get one of you lucky forum members a chance to get to the US nationals, and making it actually happen.

Give it to whoever gets the most upvotes in the meme thread?

I wouldn't expect the strongest player to win even a well-run highly-competition-centric tournament a huge percentage of the time. (Not to mention the fact that such a small percentage of the iso leaderboard is based on 3p games...)

I just really don't understand the point of a tournament where rewarding stronger play is unimportant.

[Captain_Frisk]

I suspect that the answer here is to run it like a face to face tournament.  Set a series of dates and times  (example: this Saturday and Sunday from 1pm until 5pm eastern)

We should be able to figure out how long a standard 3p game is, and look to play some # each day....

Lets say that you play in groups of 3 - randomly assigned, 3 times each (to somewhat minimize 1p advantage).  I'm going to guess that you should be able to play those 3 games in about an hour.

Play 4 sets of 3 games during this 4 hour window, and total up win points for the day (say, Winner gets 3 points, 2nd place gets 1, Loser gets nothing.... ties split the points)

At the end of Saturday, draw a line to cut that gets you down to 9 players to compete on Sunday - with alternates if need be.

On Sunday, same format, but this time everyone plays everyone else 3 times.

P1,P2,P3,P4,P5,P6,P7,P8,P9

Round 1
P1,2,3    P4,5,6   P7,8,9

Round 2
P1,4,5   P2,6,7    P3,8,9

Round 3
P1,6,9  P2,8,4  P3,7,5

Round 4
P1,7,8  P2,5,6   P3,4,9

Slate is wiped clean on Sunday - and most points wins.  Ties go to standings from Saturday.  Ties on that go to a 2p playoff best of 5, or on Isorating, or on who gives theory the most money.

[pingpongsam]

I just really don't understand the point of a tournament where rewarding stronger play is unimportant.

I'm going to agree with this sentiment with the reservation that it certainly gives me a much higher probability of actually sneaking in there.

[Captain_Frisk]

Or, force everyone to play non stop games with KC, Possession, Scrying Pool, Golem, Worker Village, Quarry, Throne Room, Masquerade, Spy, Secret Chamber, Platinum and Colony

The last one who hasn't broken their computer, walked away or rage quit wins.

[theory]

I just really don't understand the point of a tournament where rewarding stronger play is unimportant.

I'm going to agree with this sentiment with the reservation that it certainly gives me a much higher probability of actually sneaking in there.

I think what rrenaud means is that if you define "best player" to be "the player on top after an infinitely large sample of games", then with such a small sample it is unlikely that the final player on top will be the same as if you had an infinite sample.

[WanderingWinder]

I just really don't understand the point of a tournament where rewarding stronger play is unimportant.

I'm going to agree with this sentiment with the reservation that it certainly gives me a much higher probability of actually sneaking in there.

I think what rrenaud means is that if you define "best player" to be "the player on top after an infinitely large sample of games", then with such a small sample it is unlikely that the final player on top will be the same as if you had an infinite sample.

Well sure - this is going to be true with any finite number of games, to some extent anyway. This doesn't mean you can't try to do the best you can, though.

[shark_bait]

I'll agree with Frisk, his idea seems like a great idea.  I would agree that three 3-player games easily finish within an hour.  Barring 3 consecutive slogfest style games, 30-45 minutes is a reasonable amount of time.  I like this style and think it would work in the quick nature and also provide an opportunity for good play to shine.  Although a 12-game final isn't perfect, it certainly rewards better play much more so than every other qualifier tournament.

[Captain_Frisk]

I'll agree with Frisk, his idea seems like a great idea.  I would agree that three 3-player games easily finish within an hour.  Barring 3 consecutive slogfest style games, 30-45 minutes is a reasonable amount of time.  I like this style and think it would work in the quick nature and also provide an opportunity for good play to shine.  Although a 12-game final isn't perfect, it certainly rewards better play much more so than every other qualifier tournament.

If you drop to 7 qualifiers, you can get down to a 9 game final where eveyone plays everyone else (although I'm not sure if you can get the same symmetry around not having the same pairings more than once.)

I don't really see it as a final so much as a 2nd day.  At the end of the day, you're talking about 24 games of dominion.  I'm sure most of the people who are serious about playing have played that many on a work day, let alone over a weekend.

[theory]

I just really don't understand the point of a tournament where rewarding stronger play is unimportant.

I'm going to agree with this sentiment with the reservation that it certainly gives me a much higher probability of actually sneaking in there.

I think what rrenaud means is that if you define "best player" to be "the player on top after an infinitely large sample of games", then with such a small sample it is unlikely that the final player on top will be the same as if you had an infinite sample.

Well sure - this is going to be true with any finite number of games, to some extent anyway. This doesn't mean you can't try to do the best you can, though.

Of course we will try, but I think rrenaud is just saying that it should not be the sole priority. 

[greatexpectations]

- 3p games are going to be harder to organize and will take linger to play. i would stick with 2p for as long as possible. if jay insists on 3p, make the finals and semis 3p but leave the knockouts as 2p.
- as WW said, make sure participants can actually make the US finals. competitors should be flexible, but within reason. the winner shouldn't just be a decent player with the most free time. split by weekday or day/night availability whenever possible.
- you might have to restrict it to f.ds regulars. it is a f.ds qualifier, so i think it is fair that f.ds members get first chance. of course, i have a bias here, but so will everyone inputting their opinion. maybe give preference similar to how the beta signup worked?
- 2 2p knockout rounds, a 3p semi, and a 3p final makes 36 players. if you whittle down the forum members to those in the US, having flexible schedules, and actually able to make the final that might be doable.

[WanderingWinder]

I don't think 2p is an option. Further, since it's 3p later, I think it really really ought to be 3p from the beginning. 2 to 3 is a different beast, and it doesn't make sense to me to mix them. You wouldn't have people compete in soccer for the intro rounds before switching to a rugby tournament, even if you like soccer better.
July 13th-14th are the dates for the US champs, in case anyone is wondering.

I still think as many rounds of play as possible checks most of the boxes we want - more games, which means more fun, and less randomness/more skill. The issue is how many can we deal with given people's schedules and the short time frame we're looking at.

Also, I think there's a bit of an overestimation of how long these games take. I'd guess 20-30 min on average.

[pingpongsam]

I don't know about you guys but I'm auto-matching 3p games for the forseeable future. The tempo is totally different from 2p.

[greatexpectations]

I don't think 2p is an option. Further, since it's 3p later, I think it really really ought to be 3p from the beginning. 2 to 3 is a different beast, and it doesn't make sense to me to mix them. You wouldn't have people compete in soccer for the intro rounds before switching to a rugby tournament, even if you like soccer better.

well obviously mixing formats is not the preferred option dude. come on now. but we don't have the time to make a nice 243 player tournament spread out over two months. there needs to be concessions somewhere, and 2p saves time and is easier to arrange matchups.

i actually think you are underestimating game lenghts. 3p is longer than 2p, and all tournament sets have always taken longer than a few games casual on isotropic would. combining this leads to long games, and you just can't do that when trying to fit the schedule of 3p. it seems ridiculous to me that the tournament would be so heavily dependent on who has the most free time to play.

[Young Nick]

I think that one of the first things to do is make an official sign-up. Set a final date that people must sign up by and hold them to it. Once we have an idea of how many people we are talking about, things become a lot easier. It could be that for a time-intensive tournament, only 30 people are interested. Hell, look at Swiss Isodom; there aren't that many on there and it is much more flexible in scheduling games and has a number of international competitors.

All I am saying is that 200 is a lot different than 50.

[shark_bait]

i actually think you are underestimating game lenghts. 3p is longer than 2p, and all tournament sets have always taken longer than a few games casual on isotropic would. combining this leads to long games, and you just can't do that when trying to fit the schedule of 3p. it seems ridiculous to me that the tournament would be so heavily dependent on who has the most free time to play.

Every other qualifier is dependent on where people live, that is a much bigger restriction than the "3-player games are harder to schedule" restriction.  Keep in mind that nothing has been decided yet.  I'm sure that a format can be arranged that will do 2 things.

1.)  Preserve the format to closely match the irl multiplayer set up.
2.)  Allow many people to play somewhat independent of time
      ****obviously available time will be somewhat of a factor as it is in any tournament*****

I'm very optimistic that this tournament will cater to both of these aspects.  In fact, I would hope that this tourney would be more readily available to more people than compared to other regionals which require people to give pretty much an entire saturday.

[WanderingWinder]

I don't think 2p is an option. Further, since it's 3p later, I think it really really ought to be 3p from the beginning. 2 to 3 is a different beast, and it doesn't make sense to me to mix them. You wouldn't have people compete in soccer for the intro rounds before switching to a rugby tournament, even if you like soccer better.

well obviously mixing formats is not the preferred option dude. come on now. but we don't have the time to make a nice 243 player tournament spread out over two months. there needs to be concessions somewhere, and 2p saves time and is easier to arrange matchups.

i actually think you are underestimating game lenghts. 3p is longer than 2p, and all tournament sets have always taken longer than a few games casual on isotropic would. combining this leads to long games, and you just can't do that when trying to fit the schedule of 3p. it seems ridiculous to me that the tournament would be so heavily dependent on who has the most free time to play.

If it's so obvious that it's not an option, dude, why are you advocating it? I don't think 2p actually saves much if any time, at serious downsides in terms of what I would prioritize. I'm 100% against having anything other than 3p dominion here.
Let's hold our horses on how big the thing is going to be. I seriously, seriously doubt we'd have 243 people who are a) American b) have the time to play this, c)want to, and d) are able to make it to nationals. If we do, we may be forced into something like single elimination 3p, which would be sad of course.
I really do think getting a rough number is going to be really important in figuring out which formats are possible.

[Ozle]

So, decide a date range
Then find out numbers
Then decide on format?
? ? ? ? ? ? ?
= Profit

[WanderingWinder]

Competitively, I would say the rules should be as close as possible to what we'll see in the actual US Championships. I'm pretty sure that's not going to be my favourite choice , but it seems logical. I mean this in terms of how much credit goes to 1st vs 2nd vs 3rd (equal distance b/t 1-2 and 2-3, or is it more winner-take-all?), how sets are determined (if applicable), etc.
I would hope everybody gets to play at least a few games, if possible.

[Axxle]

50p dominion.  One game.  Winner take all.

[Young Nick]

50p dominion.  One game.  Winner take all.

Better rush those Estates!

[theory]

The problem with identifying a number, and then deciding on a format, is that the format may dictate how many can come.  For instance, if we do a one-day thing, or even a three-day thing, you'd have to know that in advance in order to sign up.

169 Americans/Canadians signed up for the last DSC (which had a much smaller prize).  I'm using that as an estimate for now.

[WanderingWinder]

The problem with identifying a number, and then deciding on a format, is that the format may dictate how many can come.  For instance, if we do a one-day thing, or even a three-day thing, you'd have to know that in advance in order to sign up.

169 Americans/Canadians signed up for the last DSC (which had a much smaller prize).  I'm using that as an estimate for now.

Right. But we can get an upper bound by figuring out how many have interest and can make use of the prize.

[Ozle]

I would also have a cut off date on forum accounts created before you announced it....

[Young Nick]

169 Americans/Canadians signed up for the last DSC (which had a much smaller prize).  I'm using that as an estimate for now.

Well, I guess there are a lot of other factors to consider here. There are clearly several more people who are likely to see it than the last DSC, and considering there is a bigger prize, a larger percentage of people would be inclined to participate. At the same time, there are logistical matters which probably rules it out for many. I would guess there would be at most 100 Americans who will sign up.

[greatexpectations]

If it's so obvious that it's not an option, dude, why are you advocating it? I don't think 2p actually saves much if any time, at serious downsides in terms of what I would prioritize. I'm 100% against having anything other than 3p dominion here.
Let's hold our horses on how big the thing is going to be. I seriously, seriously doubt we'd have 243 people who are a) American b) have the time to play this, c)want to, and d) are able to make it to nationals. If we do, we may be forced into something like single elimination 3p, which would be sad of course.
I really do think getting a rough number is going to be really important in figuring out which formats are possible.

uh. i answered that in the section you quoted. it's not that it is an option, it just isn't preferred. i think you continue to underestimate A. how many people we will get B. how much longer 3p games with higher stakes will take and C. how easy it will be to get 3p games together. i agree that 2p is not the preferred, but i think that for at least elimination rounds it will make scheduling easier and faster.  you can feel free to disagree, but this thread was theory asking for opinions and sorry, your opinion is not the final one. as i said before, i would prefer to not see the tournament come down to simply who has the most free time.

[ednever]

At least some of the qualifying tournaments have time caps for the prelim games. If you can finish an in-person game in 40 minutes I'll bet you can finish an isotropic game a lot faster.

The Detroit tourney is doing it this way:

5 prelim rounds. 1-point per win. Half of the participants advance based on points (tie break on total vp earned). Max 50 minute games.

That takes them to a max 32 players.

Then they do single eliminatation quarter finals (terrible idea) to bring it to 10 players.
Then single elimination semi finals to bring it to 3 players.

Winner takes all in the last game.


Obviously the single elimination games are a terrible idea for any level of fairness, but the prelim method isn't bad: play a bunch of games (5+). Half of the people advance to the tournament. You could do that for a couple of rounds to bring thr numbers down.

Assuming more than 80 participants, something like this:

9am-1130am: seeding. Play five games. 81 advance seeded
1130-1pm: 27 groupings (seeded based on points from seeding round). Each plays 3 games. Once in each chair. Winner from each group advances
1-4pm: 9 groups (using earlier seeding). Play six games, twice in each seat. Top player advances.
4-7pm: 3 groups. Repeat process from last round.
7pm+ finals. 3 players, xx games to determine the champion

Assumes games are 30 minutes each. The final could be 10 games and it will still be done by midnight at the latest.

Or you could break it up and do one round a day. It would require the finalists to be available 5 days instead of one, but same principles will apply.

Fwiw I agree with WW that the games should all be three player (with the added benefit that, though they take longer and ate more random, they do eliminate 2/3rds of people per round instead of just half...)

Ed

[WanderingWinder]

If it's so obvious that it's not an option, dude, why are you advocating it? I don't think 2p actually saves much if any time, at serious downsides in terms of what I would prioritize. I'm 100% against having anything other than 3p dominion here.
Let's hold our horses on how big the thing is going to be. I seriously, seriously doubt we'd have 243 people who are a) American b) have the time to play this, c)want to, and d) are able to make it to nationals. If we do, we may be forced into something like single elimination 3p, which would be sad of course.
I really do think getting a rough number is going to be really important in figuring out which formats are possible.

uh. i answered that in the section you quoted. it's not that it is an option, it just isn't preferred. i think you continue to underestimate A. how many people we will get B. how much longer 3p games with higher stakes will take and C. how easy it will be to get 3p games together. i agree that 2p is not the preferred, but i think that for at least elimination rounds it will make scheduling easier and faster.  you can feel free to disagree, but this thread was theory asking for opinions and sorry, your opinion is not the final one. as i said before, i would prefer to not see the tournament come down to simply who has the most free time.

I'm not saying my opinion is the final one. I merely disagreed with yours. And gave reasons why I think your idea is bad. Nothing personal. Just think it's a bad idea, and certainly think it's not the only possibility.
Man, if it were the only possibility, to get things in, why would he have asked for ideas? And why would you need to suggest it.
I don't understand how you're NOT saying this is the preferred option giving the circumstances though. So maybe I'm misunderstanding you? Because it seems pretty clear to me that you're pushing for some 2p in all your suggestions, even as in your non-concrete fluff, you're saying 'oh no, this is bad'. They don't seem to jive to me.
And with all due respect, I have no reason to see why you should have any better than the absolute vaguest idea how long 3p isotropic games should last. Add to that that you've not actually given supporting reasons, and I'm not inclined to buy your arguments on how long the games will take. Again, nothing personal against you, I just don't think you're right at all here, and you've given me no reasons to make me think you know something I don't here.

[WanderingWinder]

...sample proposal...

I generally like the look of things, but if I had my druthers, I would redistribute some of the prelim stuff into making that 11:30-1 thing.

Other, larger problems: The time-frame looks good, except that 9 am on the east coast is 6 am on the west coast, and 7 pm on the west coast is 10 on the east coast. Also, you have got to give people some time for breaks. Particularly to eat meals.
It's going to be a little tough to ensure who sits where on isotropic. Also, having person X or Y always behind/ahead of you in play can be somewhat of a strategic advantage/disadvantage.
If you throw the seating thing away, you no longer need to have a divisible-by-three number of games, which gives you some more flexibility on round length.

Finally, how do you determine who did best? Is it just most wins? Surely you don't want total VP? Is it a 2/1/0 for 1st/2nd/3rd? Or should winning mean more than that?

All that said, I generally like it, if we determine that one-day is the best way to go.

[theory]

I would also have a cut off date on forum accounts created before you announced it....

This won't happen.  A lot of people registered for the forum for the 2011 DSC and became active members afterwards.  And I see it more as an "internet qualifier" rather than a "DS qualifier".

There's a huge proportion of people that read the site / lurk the forums without registering.

[HiWay2Hello]

Finally, how do you determine who did best? Is it just most wins? Surely you don't want total VP? Is it a 2/1/0 for 1st/2nd/3rd? Or should winning mean more than that?

Somewhere along the way, there ought to be a tiebreaker that takes into account seating position, assuming this can't be fully equalized.  For example, if Player A wins 3 out 5, but his seat order was 1/2/1/2/1, all else equal he should be ranked lower than Player B who has an identical record but whose seat order was 2/3/2/3/2.  Perhaps something like "tiebreak" points could be given, with 3 points for every game in third seat, 2 points for every game in second seat, and 1 point for every game in first seat.

[greatexpectations]

And with all due respect, I have no reason to see why you should have any better than the absolute vaguest idea how long 3p isotropic games should last. Add to that that you've not actually given supporting reasons, and I'm not inclined to buy your arguments on how long the games will take. Again, nothing personal against you, I just don't think you're right at all here, and you've given me no reasons to make me think you know something I don't here.

ok, a few problems here.  first off, you call me out for having no evidence but you provide non yourself.  not sure that is how things work. it seems to me that the reasons why 3p games are longer than 2p games are actually painfully obvious.  more players, more analysis paralysis, more reactions to attacks, attacks generally having more power, more provinces/colonies to clear, trickier end game calculations to manage. the only case i see for 3p games being faster is that piles can run faster.  but again, with 3p that is something you need to be more aware of anyway and play might be tailored to avoid that situation.

furthermore, you completely ignore the other half of my reasoning - 3p games are harder to arrange.  i have not played any 3p tournament games on here, but i will open it up to all the players who played multiplayer in the Isodom: Challenge. was it easier or harder than in 2p tournaments to set up your game dates.  i seem to recall a thread which indicated that arranging multiplayer matches was difficult. now lets that expand that past a single case of just a few high level players who play far more than average in the first place to potentially hundreds of players.

and you say that i shouldn't have the vaguest idea of how long 3p games will take?  what are you basing that off of exactly?  my 3p game total under this nick? i can freely provide you with both of my alternate user names if you would like, and you can count up my 3p games there.  likewise i can point you to friends on isotropic who i have played with IRL.  your opinion on whether to buy a gold or a duchy turn 10 in a BM game might be more valid than mine, but i see no reason to believe why the same should hold with our opinions on 3p games.

[greatexpectations]

a few things to consider:
- forum membership and isotropic users have both increased drastically since the DS.com championships.  we could easily see more than the DS.com tournament.
- there does not seem to be any reason why a player would not sign up.  if they miss a match and forfeit, oh well. there is no conceivable punishments for dropouts. this tournament seems to offer some level of prestige as well as a very nice physical prize.  there is no reason why people who might be pressed for time might not start, see how it goes for them, and carry on from there. 
- collusion? kingmaking? tiebreakers? higher stakes and anonymity could cause trouble.

[ednever]

I generally like the look of things, but if I had my druthers, I would redistribute some of the prelim stuff into making that 11:30-1 thing.
[/quote]

I don't follow. Do you mean you would increased the 1130-1pm slot, or decrease it? Or something else?

Other, larger problems: The time-frame looks good, except that 9 am on the east coast is 6 am on the west coast, and 7 pm on the west coast is 10 on the east coast. Also, you have got to give people some time for breaks. Particularly to eat meals.

True. The last one day tourney started at 6am Pacific Time, so worse case you do that. You could split the difference - start at 7 or 8am PT and have the finals start at 8 or 9pm Eastern Time. Or even have the finals happen on a separate day.

It's going to be a little tough to ensure who sits where on isotropic. Also, having person X or Y always behind/ahead of you in play can be somewhat of a strategic advantage/disadvantage.
If you throw the seating thing away, you no longer need to have a divisible-by-three number of games, which gives you some more flexibility on round length.

Makes sense. Easiest way would be to use the isotropic default which will give preferred seating to whoever lost the previous round (I have no idea what iso does with the person who came second? Would it just reverse the order. If three players came in 1-2-3, would their order be 3-2-1 automatically in the next round?

Finally, how do you determine who did best? Is it just most wins? Surely you don't want total VP? Is it a 2/1/0 for 1st/2nd/3rd? Or should winning mean more than that?

I thought I would leave that to someone else. I think the three obvious options are:
(1) 1 point for a win, nothing for 2nd/3rd
(2) 2 points for a win, 1 point for 2nd (i.e. 2-2 ties with 1-3)
(3) 3 points for a win, 1 point for 2nd (i.e., 2-2-2 ties with 1-3-3)

I don't like total VP for a tie breaker. Maybe you use seating order for a tie breaker - as is seating order in the first game, or 'seating order points (same as the points above for seating instead of winning, but higher number is bad for the purposes of tie breaking only)

All that said, I generally like it, if we determine that one-day is the best way to go.

Same type of thing works for multi-day. Just do 1+ round per day instead of all on a single day.

Ed

[WanderingWinder]

And with all due respect, I have no reason to see why you should have any better than the absolute vaguest idea how long 3p isotropic games should last. Add to that that you've not actually given supporting reasons, and I'm not inclined to buy your arguments on how long the games will take. Again, nothing personal against you, I just don't think you're right at all here, and you've given me no reasons to make me think you know something I don't here.

ok, a few problems here.  first off, you call me out for having no evidence but you provide non yourself.  not sure that is how things work. it seems to me that the reasons why 3p games are longer than 2p games are actually painfully obvious.  more players, more analysis paralysis, more reactions to attacks, attacks generally having more power, more provinces/colonies to clear, trickier end game calculations to manage. the only case i see for 3p games being faster is that piles can run faster.  but again, with 3p that is something you need to be more aware of anyway and play might be tailored to avoid that situation.

furthermore, you completely ignore the other half of my reasoning - 3p games are harder to arrange.  i have not played any 3p tournament games on here, but i will open it up to all the players who played multiplayer in the Isodom: Challenge. was it easier or harder than in 2p tournaments to set up your game dates.  i seem to recall a thread which indicated that arranging multiplayer matches was difficult. now lets that expand that past a single case of just a few high level players who play far more than average in the first place to potentially hundreds of players.

and you say that i shouldn't have the vaguest idea of how long 3p games will take?  what are you basing that off of exactly?  my 3p game total under this nick? i can freely provide you with both of my alternate user names if you would like, and you can count up my 3p games there.  likewise i can point you to friends on isotropic who i have played with IRL.  your opinion on whether to buy a gold or a duchy turn 10 in a BM game might be more valid than mine, but i see no reason to believe why the same should hold with our opinions on 3p games.

I have no evidence except my experience. I have played a decent amount of multiplayer (not a ton, but a decent amount). CR says you've played... one game. IRL games take way longer, of course, but that's irrelevant. And games with your friends tend to take longer too, but again, not relevant. 3p games take longer of course, but given that you need only 2/3 as many, it seems to roughly cancel out, from my experience. Actually, it seems a little faster, because mega-turns still mega, and 3-piles come sooner. If you get attacked to death, it can detract some, but... well, this is at least cancelled out.
As for scheduling, first of all, I WAS one of the people in isodom challenge. It wasn't noticeably more difficult to set things up. Moreover, if you're thinking this is going to be a 'here are your opponents, figure out when is good for you thing'... well, it won't be. There isn't time. So, given that it's going to be pre-arranged day(s) X, time(s) Y, finding games shouldn't be too hard. It's pretty hard to get multiplayer games of decently-leveled people on iso now, because most people play 2p. But with the backdrop of the tournament, it shouldn't be difficult at all.

[zxcvbn2]

A couple things I have to say. I generally like Ed's format, and would like to throw my hat in the ring for a multiple-day tournament, as opposed to one day, because it allows for us to play more games per round if we want to. Everything's really limited if we force it into 12 hours worth of playing time or so.

Also, is it possible we could have a scoring system based on what player wins? As in maybe if first player wins he gets 3 points, 2nd player gets 4, and 3rd player gets 5? Those are just arbitrary numbers, and I think it would work best if the winner got all the points, though I am certainly no expert on these things. Anyway, what do you guys think about that?

[WanderingWinder]

A couple things I have to say. I generally like Ed's format, and would like to throw my hat in the ring for a multiple-day tournament, as opposed to one day, because it allows for us to play more games per round if we want to. Everything's really limited if we force it into 12 hours worth of playing time or so.

Also, is it possible we could have a scoring system based on what player wins? As in maybe if first player wins he gets 3 points, 2nd player gets 4, and 3rd player gets 5? Those are just arbitrary numbers, and I think it would work best if the winner got all the points, though I am certainly no expert on these things. Anyway, what do you guys think about that?

I dislike the idea. A lot. 1p is an advantage, and moreso in multiplayer. But not to THAT extent.

[zxcvbn2]

A couple things I have to say. I generally like Ed's format, and would like to throw my hat in the ring for a multiple-day tournament, as opposed to one day, because it allows for us to play more games per round if we want to. Everything's really limited if we force it into 12 hours worth of playing time or so.

Also, is it possible we could have a scoring system based on what player wins? As in maybe if first player wins he gets 3 points, 2nd player gets 4, and 3rd player gets 5? Those are just arbitrary numbers, and I think it would work best if the winner got all the points, though I am certainly no expert on these things. Anyway, what do you guys think about that?

I dislike the idea. A lot. 1p is an advantage, and moreso in multiplayer. But not to THAT extent.

Fair enough. Just out of curiosity, would you like it if there was a smaller difference between the numbers (like 3-3.25-3.5 or something smaller), or are you against it entirely?

[WanderingWinder]

A couple things I have to say. I generally like Ed's format, and would like to throw my hat in the ring for a multiple-day tournament, as opposed to one day, because it allows for us to play more games per round if we want to. Everything's really limited if we force it into 12 hours worth of playing time or so.

Also, is it possible we could have a scoring system based on what player wins? As in maybe if first player wins he gets 3 points, 2nd player gets 4, and 3rd player gets 5? Those are just arbitrary numbers, and I think it would work best if the winner got all the points, though I am certainly no expert on these things. Anyway, what do you guys think about that?

I dislike the idea. A lot. 1p is an advantage, and moreso in multiplayer. But not to THAT extent.

Fair enough. Just out of curiosity, would you like it if there was a smaller difference between the numbers (like 3-3.25-3.5 or something smaller), or are you against it entirely?

I'm against it entirely. Probably, if there's a small enough discrepancy in the numbers, you could make it balanced. But it just seems weird to me. A win should be a win should be a win, even if it's not entirely fair between the players. Especially because 1st-turn advantage moves a lot from kingdom to kingdom, and man, you do NOT want to have to control for that.

[zxcvbn2]

Okay, yeah, I understand, I just wonder if there could be a quick fix to the turn order problem since isotropic doesn't allow you to designate turn order. I suppose you could just all sign out and keep proposing games until you get the correct turn order instead.

[Kirian]

I wonder if Jay realizes at all how much easier 2P tournaments are vs 3P, just from a tournament logistics standpoint?  This would be true even in an IRL tournament; I'd love to know how they'll run Nationals and Worlds.

For the sake of putting matches together, obviously this has to happen on the same day or days.  Weekdays are likely a no-go; if you do "Tuesday evening," you can't reasonably start before 5:30 PM Pacific, and can't reasonably end after, say, midnight Eastern--a 4-hour block isn't going to be enough to do much.

I'm going to cautiously advocate a "Swiss-like" qualifier with a semifinal elimination round and longer finals.  Format as follows:

Qualifier:

Individual three-game matches between three players.  Within the match, players receive 2 pts for 1st, 1 for 2nd, 0 for 3rd; ties split all points.  Points from the three games are totaled, and those scores reported to the hosts.

Match points are given as 2-1-0, again with points split for ties.  Triplets are formed so as to match players who have equal numbers of points.  Alas, this will have to be hand-calculated, but I know we have a few people here who are good with Google spreadsheets (and who, you know, run the place), so I wouldn't worry too much about that.  This creates an ordering of any field <243 within five rounds--about 8 hours including food breaks, calculations time, etc.  Sure, it's not a perfect ordering, but it's enough to create a smaller elimination field of 27.  If we get >243 players, well, tack on an extra round there.

Swiss tie-breaker:  sum of player positions over the 15 or 18 games played.  This acknowledges the importance of position in the same way Dominion itself does, by using it as a tiebreaker.  Second tie-breaker by whatever standard Swiss-style tiebreaker you wish.

Day two, three elimination rounds.  Three games in the first, five in the second, seven in the finals.  About 4 hours.

In fact, maybe better to play 6 Swiss rounds anyway, 3 each on a Saturday and a Sunday, then eliminations the following weekend.

Anyway, my two cents.

[theory]

I do not think it is true in real life that 2P is easier than 3P.

Keep in mind that Nationals/Worlds are actually 4P.  I lobbied to have our tournament as 3P, solely for organizational convenience, but I can be persuaded to alter it to 4P if people think that is better for a qualifier.  Running 2P IRL requires twice as many sets, twice as much space, and twice as many matches.

I seriously thought about sponsoring an IRL 2P Dominion tournament, but if you try to work out the logistics of it just for a couple minutes you will rapidly discover how brutal it is.  The easiest way to run an in-person 2P tournament is to hand everyone a laptop and tell them to sign onto Isotropic.

[WanderingWinder]

I think 3p is a much better format for dominion, generally, than 4p. The seating, the luck, the slogs, the provinces-per-player, all point to 3p generally being nicer for me.
Having said that, I would lobby for 4p if nationals and worlds are 4p. I think it makes sense for the conditions to be the same in qualifying as the later event. I'm not sure that, given that we need to probably have everything on a day (or a few days) and not the 'find a time with your opponent that works' that we're more used to, it's actually logistically easier to do 3p than 4p. At least significantly so. Of course, you're the one organizing the logistics, not me . And I do think that 4p are significantly faster per player than the others, because 3p endings move up a lot, and there are fewer provinces per player.

[zxcvbn2]

I'd basically like to agree with what Wandering Winder said, especially that 3 player and 4 player in the type of setting we're looking at are no more difficult logistically.

Also, I slightly prefer Kirian's format to all the others, though I think it would take the most work from an organizer's standpoint, and I would prefer the first round of the knockout round would be longer (Really I don't see the point in playing different size matches in each round of the knockout phase at all).That's basically all I have to add.

[zxcvbn2]

Also, if you do decide to use Kirian's format (not saying you should, but it's the most "liked " at this point so I used it), I set up a blank template you could potentially use. It's not as convenient as an online generator like challonge (which, if anyone knows of one that will do a swiss for 3-player matches, please let me/us know) and will require a lot more calculation on the part of whoever runs it (which I would love to help if that's needed or wanted).

Anyway, this is what I've got: https://docs.google.com/spreadsheet/ccc?key=0Ah6dTyQTfYBBdDZMc1BVam9BN2hkWG1VZ0VwRFZkOWc

Feel free to use and edit this if that's the format you decide to go with.

[yuma]

I don't think anyone has mentioned this, but Kirian do you think we should postpone IsoDom 5 during this tournament? I know we are racing against an invisible clock with Isotropic going down at some point, but we don't have a strict timeline to compete against. What do you think and what do others competing in IsoDom 5 think?

[Kirian]

I think 3p is a much better format for dominion, generally, than 4p. The seating, the luck, the slogs, the provinces-per-player, all point to 3p generally being nicer for me.
Having said that, I would lobby for 4p if nationals and worlds are 4p. I think it makes sense for the conditions to be the same in qualifying as the later event. I'm not sure that, given that we need to probably have everything on a day (or a few days) and not the 'find a time with your opponent that works' that we're more used to, it's actually logistically easier to do 3p than 4p. At least significantly so. Of course, you're the one organizing the logistics, not me . And I do think that 4p are significantly faster per player than the others, because 3p endings move up a lot, and there are fewer provinces per player.

I'll agree with WW here that if nationals are 4P, the logistics for 4P vs. 3P on Iso shouldn't be different at all, and 4P would be matched with the nationals.

Yuma, I suspect a postponement will definitely be in order.

[theory]

Thank you all for your discussion.  It was extremely helpful.  Unfortunately, we don't think we can run a swiss-style system, or heats in the preliminary qualifiers.  Instead, what we'll do is this:

We will run a 4-player qualifying tournament.  We will have four Qualifying Days:

Sunday, June 24
Tuesday, June 26
Thursday, June 28
Saturday, June 30

Each day, we will run a single-elimination knockout tournament ("Knockout Qualifier").  Each Knockout Qualifier will produce 4 players that advance.  The final 16 "semifinalists" will duke it out amongst themselves into 4 finalists, and then to 1 champion.  The Semifinals/Finals would be done sometime between June 30 and July 2, the final deadline to have a winner.

For each day, we will have registration the day before.  This is intended to minimize the risk of a no-show.  You may register for as many Knockout Qualifiers as you wish, but of course you can only advance once into the semifinals.

I'll prepare a front page post soon.  If you have any other feedback, now is the time to provide it.

[theory]

The most contentious part about this -- and the part I'm most willing to change -- is how the Knockout Qualifiers are structured.  Single-elim 4p knockout sucks.  Is it that much better if we have pre-arranged heats, based on the registrant list, and draw our top 4 from whoever has the most win points for the day?

What I'm not flexible on (and we will definitely have) is pre-registration and static brackets. 

[zxcvbn2]

The final 16 "semifinalists" will duke it out amongst themselves into 4 finalists, and then to 1 champion.  The Semifinals/Finals would be done sometime between June 30 and July 2, the final deadline to have a winner.

Will the semifinals and finals also be single elimination or will they be best-of-X? I know it's difficult scheduling those matches in such a short amount of time, but I'd love to see first to two wins, at least, or maybe most points (3-2-1-0) through three matches if we could.

I'll also put my hat in the ring for pre-arranged heats.

[WanderingWinder]

The most contentious part about this -- and the part I'm most willing to change -- is how the Knockout Qualifiers are structured.  Single-elim 4p knockout sucks.  Is it that much better if we have pre-arranged heats, based on the registrant list, and draw our top 4 from whoever has the most win points for the day?

What I'm not flexible on (and we will definitely have) is pre-registration and static brackets. 

It seems to me that the biggest limiting factor will be that you only have time for X games per player, and so to maximize the number of games each player has, yes, pre-arranged heats are that much better. i.e. if we have time for 5 rounds of knockout, it's better for each person to play 5 heats than for one non-first to totally knock you out. I suppose it's a little more work to arrange, because it's more games. But competitively, yes, it's THAT much better.

[ednever]

A compromise (between "fairness" and "expedited-ness") could be:

Each qualifying day players are put into groups of four (randomly? Seeded based on isotropic rank?)

That group of four plays xx games (4? 8? 12?). The winner by points of those games advances. (And lots of ways to determine points... 1-0-0-0, 2-1-0-0, 3-2-1-0, 4-2-1-0, 10-5-1-0, etc... Is there a way they are doing it at Nationals?)

You would only need to play (daily entrants)/4 rounds on a qualifying day. I doubt there will be more than 64 entrants on any given day - so three rounds. At 4 games/round, that's a max of 12 games.


Question:
What times are you planning to do it Theory? During the day or evenings? (Pacific vs Eastern would be good to understand too)

Ed

[theory]

Let's suppose the format alters to this:

On each day, we take the X registrants.  Divide them into X/4 groups: Group 1, Group 2, Group 3, etc.

Each group plays 4 games against itself, rearranging their seat order each time.

Award 4 points for win, 3 for second, etc.

Total up the score.  From that day's results, pick the top 4 people.

Statistically, is that enough to meaningfully take the top 4 people?  (Keeping in mind that if you are unlucky you can qualify on another day.)  Or do we need a Stage 2 / Stage 3, as Ed suggests?

[RisingJaguar]

Each group plays 4 games against itself, rearranging their seat order each time.

One thing I remember from Isodom Semifinals, I have no idea how to the seating works for 4P aside from Winner = Last.  Thus it would be hard to get that perfectly balanced (unless I'm missing something). 

[ednever]

Theory- Not a bad idea.

Here are the two key challenges to that model I can think of:
(1) If two strong players are in the same group, not only is it harder to qualify (because if you are one of those guys you are playing someone else good), but it likely means NEITHER of them qualify - as they both lower each other's point totals

(2) If you come in fourth your first game chances are you will not qualify (depending on the numbers). At that point you are just a spoiler trying to stop the top person you are playing against from advancing. There may even be drop outs (or more likely just weak playing causing more randomness)

-=-=-=-
I like the idea of the winner of each 'heat' advancing - especially in a comparison setting like Dominion (if it was something like track they often take the top in each heat + the best times, and there people running next to you are a lot less important)

I think what you are trying to do is reduce the need for people to "wait" for a second heat (because with 4 advancing from each qualifying day that means you only need two rounds for up to 64 players)

What if, if there are more than 16 registered, the top 16 playoff the next day?

So:
Sun w/top 16 play on Monday
Tue w/top 16 play on Wed
Thr w/top 16 play on Fri
Sat w/top 16 play on Sun

?

Too much work? It involves double the commitment per round (if you win), but with four options people should be able to do it?

Ed

[WanderingWinder]

I would think that what you would want to do is not lock the pods in. So what you do is label everyone A1, B1, C1, D1, A2, B2, C2, D2, A3, B3, etc. Then in the first round, A1 plays against B1 C1 and D1; second round, no matter the results of the first, A1 plays against B2 C3 and D4, B1 plays against C2 D3 and A4. Third round A1 plays against B3 C5 and D7 (wrap around when necessary). There might be some issues with numbers not dividing evenly, but I think you get my idea. That way, you're not locked into opponents and necessarily competing directly with them, but you also get mucho games.

[theory]

Each group plays 4 games against itself, rearranging their seat order each time.

One thing I remember from Isodom Semifinals, I have no idea how to the seating works for 4P aside from Winner = Last.  Thus it would be hard to get that perfectly balanced (unless I'm missing something). 

The only thing you are missing is that dougz is a very nice human being who is gracious enough to help you when technical challenges arise.  (I spoke to him about this issue earlier today and he has a solution ready for us.)

[theory]

I would think that what you would want to do is not lock the pods in. So what you do is label everyone A1, B1, C1, D1, A2, B2, C2, D2, A3, B3, etc. Then in the first round, A1 plays against B1 C1 and D1; second round, no matter the results of the first, A1 plays against B2 C3 and D4, B1 plays against C2 D3 and A4. Third round A1 plays against B3 C5 and D7 (wrap around when necessary). There might be some issues with numbers not dividing evenly, but I think you get my idea. That way, you're not locked into opponents and necessarily competing directly with them, but you also get mucho games.

One issue with that is people waiting around for other people's games to finish.  But I understand your point.

[ednever]

Hard part about shifting groups every game is MUCH harder coordination.

With one group they can sit down and play x games in a row one after another.

If they switched groups after each game play would proceed at the speed of the slowest game PLUS coordination time between games.

If everyone played x games in a single group there would be a lot more games per unit time.

I'll bet version one would he much faster to play than version two:

Version 1
Play 4 games with one group. Top player moves on. Plays 4 games with second group

Version 2
Play 5 games with different assortment of 3 players.

Ed

[Captain_Frisk]

I would think that what you would want to do is not lock the pods in. So what you do is label everyone A1, B1, C1, D1, A2, B2, C2, D2, A3, B3, etc. Then in the first round, A1 plays against B1 C1 and D1; second round, no matter the results of the first, A1 plays against B2 C3 and D4, B1 plays against C2 D3 and A4. Third round A1 plays against B3 C5 and D7 (wrap around when necessary). There might be some issues with numbers not dividing evenly, but I think you get my idea. That way, you're not locked into opponents and necessarily competing directly with them, but you also get mucho games.

One issue with that is people waiting around for other people's games to finish.  But I understand your point.

You have the same problem with a real life tournament... we can all hang out in the lobby and gloat, or read Mafia IV in another window, or kick back and have an afternoon beer.

[jsh357]

Can't make this sadly; it has to be the one week I'm traveling!  Ah well, even if I qualified I'm not sure I could afford to go to the finals.

[WanderingWinder]

The other big thing that is just looming is the seating thing. Which, are we going to do something about it, do we want to, what will we do, how do we do it, etc. etc.? It's not clear to me, though I'm not sure, given how isotropic is, that there's a whole lot we CAN do.

[RisingJaguar]

The other big thing that is just looming is the seating thing. Which, are we going to do something about it, do we want to, what will we do, how do we do it, etc. etc.? It's not clear to me, though I'm not sure, given how isotropic is, that there's a whole lot we CAN do.

Each group plays 4 games against itself, rearranging their seat order each time.

One thing I remember from Isodom Semifinals, I have no idea how to the seating works for 4P aside from Winner = Last.  Thus it would be hard to get that perfectly balanced (unless I'm missing something). 

The only thing you are missing is that dougz is a very nice human being who is gracious enough to help you when technical challenges arise.  (I spoke to him about this issue earlier today and he has a solution ready for us.)

At least I'm not the only one who forget people can be nice. 

[WanderingWinder]

The other big thing that is just looming is the seating thing. Which, are we going to do something about it, do we want to, what will we do, how do we do it, etc. etc.? It's not clear to me, though I'm not sure, given how isotropic is, that there's a whole lot we CAN do.

Each group plays 4 games against itself, rearranging their seat order each time.

One thing I remember from Isodom Semifinals, I have no idea how to the seating works for 4P aside from Winner = Last.  Thus it would be hard to get that perfectly balanced (unless I'm missing something). 

The only thing you are missing is that dougz is a very nice human being who is gracious enough to help you when technical challenges arise.  (I spoke to him about this issue earlier today and he has a solution ready for us.)

At least I'm not the only one who forget people can be nice. 

Now how did I miss that?

[metzgerism]

If you need this to go fast, I think it's gotta be a single or double elimination.

Theory...seriously, just open signups and I'll handle the format. 

EDIT: Is a 243-player double-elimination tournament going to be too large? Championships were 256...

[zxcvbn2]

Another pretty obvious thing is that they won't all be 4 player to start, correct? That is, if we get 61 people, we'll have 13 four-player games and 3 three-player games. And if so, how do we score these games? If not, are we kicking people out to meet a multiple of four? Or how is that going to work?

[metzgerism]

Also, if you're having a dilemma between 4p and 3p, there is a simple fix for this:

---

Organize players into pools of 4.
Each pool plays a 4-player game. The winner advances.
The remaining players then play a 3-player game. The winner of that also advances.
The two players who did not win are either eliminated, or sent to a loser's bracket.
In case of shared victory, play a tiebreaker between those players (2-player is fine here).

---

This scenario would be ideal if you could have two qualifiers at the end.
If you choose a winner's and loser's bracket option AND you need ONE winner, the final should go something like this:

2 players from the winner's bracket and 2 players from the loser's bracket go to the finals.
Do the format above, but if a winner's bracket player wins the 4-player game, the tournament ends and they are champion.
If a loser's bracket player wins, another 3-player game is played with that player and the two winner's bracket players, and the winner of that game is champion.

EDIT: Just read Kirian's idea...remember, I never did institute a point system in BGGDL, and wouldn't advocate for one here either. With such short notice, I think just having a knockout tournament of some form (and the above is technically a pseudo-quadruple elimination) will be much simpler, and capping at a certain number of players as well.

[metzgerism]

Another pretty obvious thing is that they won't all be 4 player to start, correct? That is, if we get 61 people, we'll have 13 four-player games and 3 three-player games. And if so, how do we score these games? If not, are we kicking people out to meet a multiple of four? Or how is that going to work?

It's probably too short of notice to get a good "soft cap" format in. At a certain point, the tournament should be capped.

[theory]

If we have:

12 registrants we do 3x 4p
13: we do 3x 3p + 1x 4p
14: 2x 3p + 1x 4p
15: 1x 3p + 3x 4p

And so on.  It shouldn't be an issue.

[theory]

I would think that what you would want to do is not lock the pods in. So what you do is label everyone A1, B1, C1, D1, A2, B2, C2, D2, A3, B3, etc. Then in the first round, A1 plays against B1 C1 and D1; second round, no matter the results of the first, A1 plays against B2 C3 and D4, B1 plays against C2 D3 and A4. Third round A1 plays against B3 C5 and D7 (wrap around when necessary). There might be some issues with numbers not dividing evenly, but I think you get my idea. That way, you're not locked into opponents and necessarily competing directly with them, but you also get mucho games.

What is the drawback of this other than waiting around for your opponent?  This is what I am leaning towards now.

To be clear:

We get a list of X registrants for Day 1

We assign everyone 4 matches, ensuring everyone has fair seating order through the day

Match 1 starts at (let's say, this is randomly chosen) 2PM EST
Match 2 at 2:30PM EST
Match 3 at 3:00PM EST
Match 4 at 3:30PM EST

10 points for winner
5 points for second place
3 points for third
1 point for fourth

In the event of a 3p game, it is still 10-5-3. 

Top 4 players of the day are chosen.

Ties broken on sum of opponent strength (so 3p gamers lose out on tiebreaks) and performance in fourth position.

[zxcvbn2]

If we have:

12 registrants we do 3x 4p
13: we do 3x 3p + 1x 4p
14: 2x 3p + 1x 4p
15: 1x 3p + 3x 4p

And so on.  It shouldn't be an issue.

Right. My only question is: How would we score a 3p game vs a 4p game?

[Kirian]

I would think that what you would want to do is not lock the pods in. So what you do is label everyone A1, B1, C1, D1, A2, B2, C2, D2, A3, B3, etc. Then in the first round, A1 plays against B1 C1 and D1; second round, no matter the results of the first, A1 plays against B2 C3 and D4, B1 plays against C2 D3 and A4. Third round A1 plays against B3 C5 and D7 (wrap around when necessary). There might be some issues with numbers not dividing evenly, but I think you get my idea. That way, you're not locked into opponents and necessarily competing directly with them, but you also get mucho games.

What is the drawback of this other than waiting around for your opponent?  This is what I am leaning towards now.

I can't think of one.

[theory]

10-5-3-1 vs 4-3-2-1 depends on who you think should advance:

4 second places
1 first place, 3 fourth places

I think 4-3-2-1 makes more sense.

[metzgerism]

I still vote wholeheartedly against having a point system at all, and giving only winners any credit.

From a competition standpoint, it prevents odd kingmaking and collusion scenarios.
From an organizational one, the book-keeping is much simpler.

[metzgerism]

Ok, here's a concept:

Remember the rules I had set up for BGGDL for multi-player games, before we went to 2p only?
Do that, but segregate everyone into some moderate sized divisions (probably 10-12 per division).
Restrict play to 4p and 3p games, and advance 3n+1 players from the pools to the knockout stage, with a bye for division champs.

(BGGDL's format was based on wins, assuming 2p games. When you played a 3p/4p game, it was treated as having played 2/3 2p games at the same time. Winning only mattered, so if you shared victory or both lost, you could 'replay' them for a result. Ended up still working like a standard RR table).

[RisingJaguar]

I still vote wholeheartedly against having a point system at all, and giving only winners any credit.

From a competition standpoint, it prevents odd kingmaking and collusion scenarios.
From an organizational one, the book-keeping is much simpler.

But does this really prevent collusion and Kingmaking?  I would think having the point system encourages people to always play for themselves, which is what you want.   

The point system also rewards consistent play though. 

FYI I can't participate, just throwing pennies. 

[theory]

In practice I doubt it will matter, points vs wins.

Let's replace it with wins, and second places break ties.  Functionally equivalent.

[metzgerism]

I still vote wholeheartedly against having a point system at all, and giving only winners any credit.

From a competition standpoint, it prevents odd kingmaking and collusion scenarios.
From an organizational one, the book-keeping is much simpler.

But does this really prevent collusion and Kingmaking?  I would think having the point system encourages people to always play for themselves, which is what you want.   

The point system also rewards consistent play though. 

FYI I can't participate, just throwing pennies.

In a winner-take-all scenario, a player in second place is more likely to stall the game so that they have a chance to win. Because winning a game is usually all that matters in a bracket, this mirrors a bracket decently.

In a point system, a player in second place and a chance to end the game might do so, if they believe that they will lose positioning if the game continues. This is kingmaking for the first player. Because winning a game is usually all that matters in a bracket, this is particularly odd to give an incentive for.

You might get some nice participation points, but later on when you get to the "you must win" part of the tournament, you might have the WRONG type of consistency: more players that were good at rushing the end of the game with a nominal score than those who are actually good at winning a game.

[Rabid]

Surely winner takes all just promotes high variance strategies?
Ie treasure map without support.

Also from 4th seat I would consider finishing 2nd a "win".

[ednever]

I've been playing a few 3- and 4-player games this week.

My vote is definitely for points (vs wins). There are many times when the spread is such that as a third player you have no hope to catch the first player, but you do have a chance to catch the second player. With points you will keep playing in self-interest. Without points you don't care and, your only incentive left is: (1) Help the player you think is your least competition overall, or (2) Just end the game so you stop wasting your time.

This is even more true in 4-player (where there could be a simultaneous batter for 1-2 and 3-4 at the same time). Without points both players 3 and 4 become kingmakers instead of playing for themselves.

The point impacts do not need to be significant (although I think it would be better if they do), but I think then need to be there (i.e., I would prefer 10-2-1-0 to 1-0-0-0, but I think something like 4-3-2-1 or 3-2-1 or 4-2-1 would be more reasonable. I know I would be more worried facing someone who came 2nd in four games than someone who came first in one game and fourth is three others.)

-=-=-
The only case where I think points leads to a debatable decision making vs. wins is when a player is in second place with little hope of making first (but some hope), ends it to guarantee his second place finish.

I personally think this is part of playing the odds in Dominion and shows more skill on "when to end" vs just going balls out to win without any consideration of your chances. But I can see both sides.

I just think the earlier issues trump this debatable one anyway.


Ed

[ednever]

One more thought:

In the case of some games being 3-player and some being 4-player, I've seen models like this used before:

               4-player           3-player
Winner      6                      6
2nd          4                      3
3rd           2                      0
4th           0                    n/a

Recognizing that a 3rd place finish in a 4-player game is a much higher achievement than a 3rd place finish in a 3-player game. 2nd place in for player is marginally more impressive than 2nd in 3-player. Winning in each is the same.

Ed

[zxcvbn2]

Either way, is a win in a three-player match going to be equal to a win in a four-player match?

I agree with Ed. I think something like 8-3-1-0 and 6-2-0 sounds about right, though if you value winning more, you could make it 10-3-1-0 and 8-2-0.

[metzgerism]

I've been playing a few 3- and 4-player games this week.

My vote is definitely for points (vs wins). There are many times when the spread is such that as a third player you have no hope to catch the first player, but you do have a chance to catch the second player. With points you will keep playing in self-interest. Without points you don't care and, your only incentive left is: (1) Help the player you think is your least competition overall, or (2) Just end the game so you stop wasting your time.

This is even more true in 4-player (where there could be a simultaneous batter for 1-2 and 3-4 at the same time). Without points both players 3 and 4 become kingmakers instead of playing for themselves.

The point impacts do not need to be significant (although I think it would be better if they do), but I think then need to be there (i.e., I would prefer 10-2-1-0 to 1-0-0-0, but I think something like 4-3-2-1 or 3-2-1 or 4-2-1 would be more reasonable. I know I would be more worried facing someone who came 2nd in four games than someone who came first in one game and fourth is three others.)

-=-=-
The only case where I think points leads to a debatable decision making vs. wins is when a player is in second place with little hope of making first (but some hope), ends it to guarantee his second place finish.

I personally think this is part of playing the odds in Dominion and shows more skill on "when to end" vs just going balls out to win without any consideration of your chances. But I can see both sides.

I just think the earlier issues trump this debatable one anyway.


Ed

I disagree wholeheartedly

This tournament has a couple issues you might be forgetting:

1. We don't have time to contemplate point systems;
2. Organizing with a point system might be a total mess for whoever admininstrates; and,
3. This is not a participation tournament - this is a tournament for WINNING.

As a player, sure you might FEEL better getting in 2nd place in a game instead of 4th place...but in a tournament setting THERE SHOULD BE NO DIFFERENCE. This isn't a "simulate game night" tournament, this is a "find the best 4p Dominion player" tournament, and a point system gets in the way of that while also making it complicated to organize.

*insert old-timey "in my day we didn't get trophies for losing" anecdote*

---

I think a lot of you are misinterpreting what a point system will do. It WON'T let you play in your self-interest - it will make you metagame. You'll be playing the point system, and not Dominion. You'll be "playing for the draw", instead of simply trying to win the damn game. When bracket time comes around, that's not going to work, and we should not be rewarding it at the expense of simplification.

"Playing for 2nd" means nothing when first place is the only thing that's going to get you a trip to Nationals. Why should we have people good at "playing for 2nd" get the best shot at the prize?

[metzgerism]

Surely winner takes all just promotes high variance strategies?
Ie treasure map without support.

I don't understand...how is this is a bad thing when winner takes all?

EDIT: There's a reason I keep getting +1's from rrenaud, and I'm waiting for him to speak up about it

[zxcvbn2]

But if you have no chance at getting first place late in the game and have nothing to play for, then you begin to worry about kingmaking. This doesn't matter as much, as long as 2nd place finishes are used as a tiebreaker anyway.

However, I don't like that three-player and four-player wins are of equal worth. I understand we don't really have a lot of time to debate the merits of each person's idea and stuff, but winning a three-player game is a lot easier than a four-player, IMO.

[Rabid]

Surely winner takes all just promotes high variance strategies?
Ie treasure map without support.

I don't understand...how is this is a bad thing when winner takes all?

It is not a bad strategy for a player given this structure, but I don't think it is player behaviour that should be rewarded heavily.
Promoting high variance low skill strategies is going to give you a fairly random winner I think.
My definition of "best player" would include a wider range of player skills.

[metzgerism]

Surely winner takes all just promotes high variance strategies?
Ie treasure map without support.

I don't understand...how is this is a bad thing when winner takes all?

It is not a bad strategy for a player given this structure, but I don't think it is player behaviour that should be rewarded heavily.
Promoting high variance low skill strategies is going to give you a fairly random winner I think.
My definition of "best player" would include a wider range of player skills.

My definition of "best player" would stop at "wins the most." This does not include "appears to be most well-rounded."

The point about having a fairly random winner, that's more about the tournament organization and less about the debate of having a point system or not - there's a reason why we went to best 2/3 and 3/5 in BGGDL, and the DSC was 4/7.

A properly laid-out format will accurately prevent wild play from being successful. A point system will do that, too...but not accurately. Again, metagaming; players will be much more unpredictable when they have an array of results to go for (especially in the final game before a bracket), being conservative in an early game and wacky in a late one. "Winner take all" eliminates that unpredictability.

But if you have no chance at getting first place late in the game and have nothing to play for, then you begin to worry about kingmaking.

Or you could alter your strategy to take advantage of the only possible avenue left to you to win.
The big issue I see here from a theoretical standpoint is another player going non-optimal at the end game, and you're penalized for their derp. In that case, we need a format that allows recovery from that.

By the way, having a point system would make that WORSE.

However, I don't like that three-player and four-player wins are of equal worth. I understand we don't really have a lot of time to debate the merits of each person's idea and stuff, but winning a three-player game is a lot easier than a four-player, IMO.

You're jumping to that conclusion. Don't worry - if there's an inequity, it will be mitigated somehow.

[ednever]

I disagree wholeheartedly

This tournament has a couple issues you might be forgetting:

1. We don't have time to contemplate point systems;
2. Organizing with a point system might be a total mess for whoever admininstrates; and,
3. This is not a participation tournament - this is a tournament for WINNING.

As a player, sure you might FEEL better getting in 2nd place in a game instead of 4th place...but in a tournament setting THERE SHOULD BE NO DIFFERENCE. This isn't a "simulate game night" tournament, this is a "find the best 4p Dominion player" tournament, and a point system gets in the way of that while also making it complicated to organize.

*insert old-timey "in my day we didn't get trophies for losing" anecdote*

---

I think a lot of you are misinterpreting what a point system will do. It WON'T let you play in your self-interest - it will make you metagame. You'll be playing the point system, and not Dominion. You'll be "playing for the draw", instead of simply trying to win the damn game. When bracket time comes around, that's not going to work, and we should not be rewarding it at the expense of simplification.

"Playing for 2nd" means nothing when first place is the only thing that's going to get you a trip to Nationals. Why should we have people good at "playing for 2nd" get the best shot at the prize?

Yeah. I think we definitely disagree. I'm happy with whatever Theory decides. Until he does, my two cents on your comments:

1. We don't have time to contemplate point systems

-> Agreed. Not a lot of time. But something like 3-2-1-0 doesn't take a ton of time. And I don't think it makes a ton of difference on what the system is. Theory can just make an executive decision.

2. Organizing with a point system might be a total mess for whoever admininstrates; and,

-> Might be, but I don't think it has to be. The players in a group can self-report. Even into a shared Google Spreadsheet.

3. This is not a participation tournament - this is a tournament for WINNING.

-> I guess the question is "what does winning mean". My feeling is (and I might be very wrong) is that you feel the player who wins a particular game is the "winner". I think my general philosophy is the player who performs well in a series of games is the "winner".

It goes back to, which of these two players is the "winner" in a series of four-player games:

1st, 1st, 4th, 4th, 4th, 4th, 4th, 4th
1st, 2nd, 2nd, 2nd, 2nd, 2nd, 2nd, 2nd

If no criteria were set out at the start I would vote for the second guy (sounds like you would vote for the first guy).
If the question was: Which of these two players is most likely to win the "9th game", I would bet on my guy every time.

If there was defined criteria beforehand - i.e., only first place matters. Then the first guy may indeed be the best player - playing to the criteria.

But playing so 'first place only matters' leads to some weird incentives - things like opening Treasure Map consistently, or, as fourth player playing just to spoil your competition (since you have better odds of that then you do of winning)

Ideally I like an incentive system that encourages players to just play the best they can.

Dominion has so much chance involved already, that encouraging players to increase that variance doesn't seem like a good idea to me.

(And I would hate to be neck and neck for a 1-2 finish with someone and have it decided on the whim of the 3rd/4th place player. Or be in the position as 3/4th place to be kingmaker in the game - feels lot like "where does the Robber go" that Donald talks about all the time)

Ed

[blueblimp]

One good thing about a point system is the following. A major obstacle in Dominion tournaments is lacking enough games to distinguish which player is best. So you want to get at much information from each game as possible.

If you only know the winner of a game, there are only 4 possible outcomes. If you know the finish order (1st-4th) of the players, there are 4! = 24 possible outcomes. So, 2 bits of information from winner-takes-all, and ~4.6 bits of information if you know the finish order, which is more than twice as much.

[metzgerism]

3. This is not a participation tournament - this is a tournament for WINNING.

-> I guess the question is "what does winning mean". My feeling is (and I might be very wrong) is that you feel the player who wins a particular game is the "winner". I think my general philosophy is the player who performs well in a series of games is the "winner".

It goes back to, which of these two players is the "winner" in a series of four-player games:

1st, 1st, 4th, 4th, 4th, 4th, 4th, 4th
1st, 2nd, 2nd, 2nd, 2nd, 2nd, 2nd, 2nd

If no criteria were set out at the start I would vote for the second guy (sounds like you would vote for the first guy).
If the question was: Which of these two players is most likely to win the "9th game", I would bet on my guy every time.

If there was defined criteria beforehand - i.e., only first place matters. Then the first guy may indeed be the best player - playing to the criteria.

But playing so 'first place only matters' leads to some weird incentives - things like opening Treasure Map consistently, or, as fourth player playing just to spoil your competition (since you have better odds of that then you do of winning)

Yeah, we're going to disagree.

* In the two players you mentioned, the criteria they play under is far more important than you assume. Are we playing for fun, or for keeps? With nothing defined beforehand, I would refrain from betting completely. People play different when different things are on the line.

Brush up on your PPR theory here. It may serve you well.

* I did mention that the format needs to allow for high-variance strategies, but not as much as I think you believe.

What if I went for 2 points every time I scored a touchdown, or kicked onside every kickoff, or went for it on 4th down every time? As a coach, I'd be fired. However, those things ARE done situationally, and can be used successfully.

What if, instead of lining up for the play, my entire defense started doing jumping jacks? It probably wouldn't work 99 times out of 100...BUT IT MIGHT WORK ONE TIME!

[metzgerism]

One good thing about a point system is the following. A major obstacle in Dominion tournaments is lacking enough games to distinguish which player is best. So you want to get at much information from each game as possible.

If you only know the winner of a game, there are only 4 possible outcomes. If you know the finish order (1st-4th) of the players, there are 4! = 24 possible outcomes. So, 2 bits of information from winner-takes-all, and ~4.6 bits of information if you know the finish order, which is more than twice as much.

Agreed on the tournament format issue.

Here's an alternative that keeps winner-take-all and gets you more information:

Play two games.

[blueblimp]

One good thing about a point system is the following. A major obstacle in Dominion tournaments is lacking enough games to distinguish which player is best. So you want to get at much information from each game as possible.

If you only know the winner of a game, there are only 4 possible outcomes. If you know the finish order (1st-4th) of the players, there are 4! = 24 possible outcomes. So, 2 bits of information from winner-takes-all, and ~4.6 bits of information if you know the finish order, which is more than twice as much.

Agreed on the tournament format issue.

Here's an alternative that keeps winner-take-all and gets you more information:

Play two games.

...

Are you serious? Your solution to "don't have time to play a lot of games" is... play more games?

[Rabid]

One good thing about a point system is the following. A major obstacle in Dominion tournaments is lacking enough games to distinguish which player is best. So you want to get at much information from each game as possible.

If you only know the winner of a game, there are only 4 possible outcomes. If you know the finish order (1st-4th) of the players, there are 4! = 24 possible outcomes. So, 2 bits of information from winner-takes-all, and ~4.6 bits of information if you know the finish order, which is more than twice as much.

Agreed on the tournament format issue.

Here's an alternative that keeps winner-take-all and gets you more information:

Play two games.

Here's an alternative that gets even more information.
Play two games and use finish order.

[metzgerism]

*probably offensive statement removed*

One good thing about a point system is the following. A major obstacle in Dominion tournaments is lacking enough games to distinguish which player is best. So you want to get at much information from each game as possible.

If you only know the winner of a game, there are only 4 possible outcomes. If you know the finish order (1st-4th) of the players, there are 4! = 24 possible outcomes. So, 2 bits of information from winner-takes-all, and ~4.6 bits of information if you know the finish order, which is more than twice as much.

Agreed on the tournament format issue.

Here's an alternative that keeps winner-take-all and gets you more information:

Play two games.

...

Are you serious? Your solution to "don't have time to play a lot of games" is... play more games?

Ehhh...not exactly the equivalency I was going for. I agree with you that in-person tournaments tend to have that issue, and you want to have a lot more rotation in a short amount of time.

Tournaments and leagues on isotropic are a different beast altogether (I've run online leagues for most of the last 7 years, and BGGDL was the last one I did - and the best). The specific parameters of this tournament actually would be MORE conducive for people to stay in their 4-player groups and play multiple games, because it will be so difficult to get things organized. Hence, more results with like opposition. Simply, we will not have the limitations of time and materials that are seen at many in-person tournaments. Instead, our challenges will be sheer game organization, and data entry.

The point I was really trying to make is this: Winner take all will probably go a little bit faster, but not much. However, since we're on isotropic, playing 2-3 games is similar in length to playing 1 game in person. I believe that's not only a reasonable and fair compromise to avoid the perverse issues of a point system, but also would be preferable for players due to our tradition of not playing one-offs for major competitions.

[zxcvbn2]

Well... that escalated quickly.

[metzgerism]

Well... that escalated quickly.

I apologize. As you might be able to see...I'm trying to edit that down.

EDIT: Screw it, just removing it. It's not worth looking like a douche to everyone to make my point.
The wheels in my head sometimes take over the fingers pressing the buttons.

[metzgerism]

One good thing about a point system is the following. A major obstacle in Dominion tournaments is lacking enough games to distinguish which player is best. So you want to get at much information from each game as possible.

If you only know the winner of a game, there are only 4 possible outcomes. If you know the finish order (1st-4th) of the players, there are 4! = 24 possible outcomes. So, 2 bits of information from winner-takes-all, and ~4.6 bits of information if you know the finish order, which is more than twice as much.

Agreed on the tournament format issue.

Here's an alternative that keeps winner-take-all and gets you more information:

Play two games.

Here's an alternative that gets even more information.
Play two games and use finish order.

Finish order is fine for tiebreakers, but not for quantitative ranking criteria.
It'd be stupid to play for tiebreakers twice when you can go for broke twice and probably get a win, nullifying any tiebreakers.

---

I'm going to recommend a point system!

Win= 10,000 points
2nd = 100 points
3rd = 1 point
4th = 0 points

Is that a little more palatable?

[WanderingWinder]

*probably offensive statement removed*

One good thing about a point system is the following. A major obstacle in Dominion tournaments is lacking enough games to distinguish which player is best. So you want to get at much information from each game as possible.

If you only know the winner of a game, there are only 4 possible outcomes. If you know the finish order (1st-4th) of the players, there are 4! = 24 possible outcomes. So, 2 bits of information from winner-takes-all, and ~4.6 bits of information if you know the finish order, which is more than twice as much.

Agreed on the tournament format issue.

Here's an alternative that keeps winner-take-all and gets you more information:

Play two games.

...

Are you serious? Your solution to "don't have time to play a lot of games" is... play more games?

Ehhh...not exactly the equivalency I was going for. I agree with you that in-person tournaments tend to have that issue, and you want to have a lot more rotation in a short amount of time.

Tournaments and leagues on isotropic are a different beast altogether (I've run online leagues for most of the last 7 years, and BGGDL was the last one I did - and the best). The specific parameters of this tournament actually would be MORE conducive for people to stay in their 4-player groups and play multiple games, because it will be so difficult to get things organized. Hence, more results with like opposition. Simply, we will not have the limitations of time and materials that are seen at many in-person tournaments. Instead, our challenges will be sheer game organization, and data entry.

The point I was really trying to make is this: Winner take all will probably go a little bit faster, but not much. However, since we're on isotropic, playing 2-3 games is similar in length to playing 1 game in person. I believe that's not only a reasonable and fair compromise to avoid the perverse issues of a point system, but also would be preferable for players due to our tradition of not playing one-offs for major competitions.

So your answer to the 'we don't have enough time to do enough games' problem is 'we do have enough time to play enough games, because it's online'? Nobody's saying that it will take as long as IRL tournaments. What they're saying is, even online, we don't have enough time to play as many games as we want.

[metzgerism]

*probably offensive statement removed*

One good thing about a point system is the following. A major obstacle in Dominion tournaments is lacking enough games to distinguish which player is best. So you want to get at much information from each game as possible.

If you only know the winner of a game, there are only 4 possible outcomes. If you know the finish order (1st-4th) of the players, there are 4! = 24 possible outcomes. So, 2 bits of information from winner-takes-all, and ~4.6 bits of information if you know the finish order, which is more than twice as much.

Agreed on the tournament format issue.

Here's an alternative that keeps winner-take-all and gets you more information:

Play two games.

...

Are you serious? Your solution to "don't have time to play a lot of games" is... play more games?

Ehhh...not exactly the equivalency I was going for. I agree with you that in-person tournaments tend to have that issue, and you want to have a lot more rotation in a short amount of time.

Tournaments and leagues on isotropic are a different beast altogether (I've run online leagues for most of the last 7 years, and BGGDL was the last one I did - and the best). The specific parameters of this tournament actually would be MORE conducive for people to stay in their 4-player groups and play multiple games, because it will be so difficult to get things organized. Hence, more results with like opposition. Simply, we will not have the limitations of time and materials that are seen at many in-person tournaments. Instead, our challenges will be sheer game organization, and data entry.

The point I was really trying to make is this: Winner take all will probably go a little bit faster, but not much. However, since we're on isotropic, playing 2-3 games is similar in length to playing 1 game in person. I believe that's not only a reasonable and fair compromise to avoid the perverse issues of a point system, but also would be preferable for players due to our tradition of not playing one-offs for major competitions.

So your answer to the 'we don't have enough time to do enough games' problem is 'we do have enough time to play enough games, because it's online'? Nobody's saying that it will take as long as IRL tournaments. What they're saying is, even online, we don't have enough time to play as many games as we want.

*sigh*

I'm not really answering the "we don't have enough time" problem. Forget that for a second.
My argument is that the solution proposed (a point system) provides a worse side-effect than the problem it supposedly solves (which it doesn't, we still won't have enough time).
The alternative I proposed was just an incredibly simple way to increase the ~2 results per matchup to ~4.

I don't believe that introducing metagaming via point system is worth the benefit it provides (if any). My argument is about weighing the options. I weigh them on the side of no point system. You might not.

In addition, if we have a pool --> bracket system, the bracket will solve incremental differences between players anyways. I'm NOT saying that a point system doesn't benefit from having a bracket attached...but I'm also not recommending having a point system at all.

[WanderingWinder]

The thing is, I really don't understand how this is getting such vehement responses on both sides. In no way is one system clearly better than the other. They're just two different systems. It's like, the question is: "Is the point of the game to win, or is the point of the game to finish in as good a placement as possible?" Both are reasonable arguments, and there's no way that one is just 'clearly' better than the other, or that either of them are ridiculous. Having a point structure does not make it non-competitive, it just makes it different.
If you look at something like auto-racing, they give points for how well you finish, not just where you place. But there are other sports where all that matters is who win. Off the top of my head, I can't think of any sports where there's a field of participants (i.e., it's not head-to-head), where they don't give more credit to 2nd place than 8th place. That doesn't make such a system 'wrong' or 'unreasonable' though. Which is better than the other, is just based on what people think it should be. That's really what matters here. Which prima facie makes nobody's view ridiculous, though you can of course prefer to NOT have that.

The thing with dominion is that this makes more significant gameplay changes, probably, than you'd see in other things. Particularly in the endgame. Again, it's not that one way is right or wrong. How can you say that playing for second is wrong? How can you say that going all out for a win is wrong? Strategically, it's all down to how the tournament is set up. Which isn't clear one way or the other.

I personally advocate for a points system, but one which has disproportionate favoring to the winner. Because I think that playing for a win is an important aspect of the game, and I think you should take the risks, particularly strategically. At the same time, I feel like 2nd and 4th shouldn't get treated the same. There's also then the issue of weird endgame behaviour/collusion-y stuff. Having winner-take-all encourages this more than a point system, because where this really comes up is in situations where it's not possible for a player to win - then they can do anything to throw the game to one guy or another, with no negative repurcussions for themselves. If 3rd is better than 4th, then this would only come into play when 4th place can't possibly even get 3rd, which is going to happen a lot less than 4th being unable to grab 1st. So this is another reason I prefer some point system.

The 'it's harder to set up' thing seems to be a weak argument to me, because it's not much harder to set up, and, more importantly, the people who would be doing the setting up seem willing to do this little extra.

So I would advocate something like 6/3/1/0. There's also the issue of the odd 3-player games, which I would probably have as - everyone plays four player games, the odd people out get buys, or probably preferentially, quasi-byes, i.e. 1 person odd gets a full bye, 2 people odd play a 2-player, where (if we assume a 6/3/1/0 point system for the moment), 1st gets 6 and 2nd gets 3, 3 odd people play a 3-player where 1st gets 6, 2nd gets 3, 3rd gets 1. But it's a bit of a thorny issue - someone is getting the better end of a luck stick, no matter how you slice it.
But more than this, I advocating trying to set up whatever point system to match what will actually be at nationals as closely as possible.

[metzgerism]

The thing is, I really don't understand how this is getting such vehement responses on both sides. In no way is one system clearly better than the other. They're just two different systems. It's like, the question is: "Is the point of the game to win, or is the point of the game to finish in as good a placement as possible?" Both are reasonable arguments, and there's no way that one is just 'clearly' better than the other, or that either of them are ridiculous. Having a point structure does not make it non-competitive, it just makes it different.
If you look at something like auto-racing, they give points for how well you finish, not just where you place. But there are other sports where all that matters is who win. Off the top of my head, I can't think of any sports where there's a field of participants (i.e., it's not head-to-head), where they don't give more credit to 2nd place than 8th place. That doesn't make such a system 'wrong' or 'unreasonable' though. Which is better than the other, is just based on what people think it should be. That's really what matters here. Which prima facie makes nobody's view ridiculous, though you can of course prefer to NOT have that.

There are a couple things to note:

1) Auto Racing doesn't have traditional playoffs.
2) All sports with traditional playoffs are 1v1 affairs.

These are HUGE differences! If you were to have an entire NASCAR season of 43-car races, then at the end say "we're going to have a 16-car bracket where it's just you against one other driver," fans would revolt!

The thing with dominion is that this makes more significant gameplay changes, probably, than you'd see in other things. Particularly in the endgame. Again, it's not that one way is right or wrong. How can you say that playing for second is wrong? How can you say that going all out for a win is wrong? Strategically, it's all down to how the tournament is set up. Which isn't clear one way or the other.

I am under the assumption that there will be a decisive bracket at Nationals, as well as one in this qualifier. THIS is what drives my argument against a point system. If there's no point system when it's most important, you shouldn't have one at any other time. Changing the parameters of "success" mid-tournament IS metagaming, no matter how many leagues do it. There are many sports leagues that DO fail at this, however (the NHL being an aggravating example) - sometimes it isn't the teams who win the most games that get invited to the tournament.

And that IS wrong.

I personally advocate for a points system, but one which has disproportionate favoring to the winner. Because I think that playing for a win is an important aspect of the game, and I think you should take the risks, particularly strategically. At the same time, I feel like 2nd and 4th shouldn't get treated the same. There's also then the issue of weird endgame behaviour/collusion-y stuff. Having winner-take-all encourages this more than a point system, because where this really comes up is in situations where it's not possible for a player to win - then they can do anything to throw the game to one guy or another, with no negative repurcussions for themselves. If 3rd is better than 4th, then this would only come into play when 4th place can't possibly even get 3rd, which is going to happen a lot less than 4th being unable to grab 1st. So this is another reason I prefer some point system.

The 'it's harder to set up' thing seems to be a weak argument to me, because it's not much harder to set up, and, more importantly, the people who would be doing the setting up seem willing to do this little extra.

So I would advocate something like 6/3/1/0. There's also the issue of the odd 3-player games, which I would probably have as - everyone plays four player games, the odd people out get buys, or probably preferentially, quasi-byes, i.e. 1 person odd gets a full bye, 2 people odd play a 2-player, where (if we assume a 6/3/1/0 point system for the moment), 1st gets 6 and 2nd gets 3, 3 odd people play a 3-player where 1st gets 6, 2nd gets 3, 3rd gets 1. But it's a bit of a thorny issue - someone is getting the better end of a luck stick, no matter how you slice it.
But more than this, I advocating trying to set up whatever point system to match what will actually be at nationals as closely as possible.

Agreed with the bold, far more than any other thing that's been said in this thread.

[rrenaud]

One good thing about a point system is the following. A major obstacle in Dominion tournaments is lacking enough games to distinguish which player is best. So you want to get at much information from each game as possible.

If you only know the winner of a game, there are only 4 possible outcomes. If you know the finish order (1st-4th) of the players, there are 4! = 24 possible outcomes. So, 2 bits of information from winner-takes-all, and ~4.6 bits of information if you know the finish order, which is more than twice as much.

From a statistical perspective, I totally agree with you.

But I think the game design trumps the statistics.  I think playing to win, all else be damned, is more in the spirit of the rules.

[rrenaud]

Also, this discussion might be useful.

http://boardgamegeek.com/thread/457599/discussion-on-a-universal-tournament-format-for-do/page/1

Although it might not be, because at least no one here is proposing rewarding players for having a high score.

[metzgerism]

One good thing about a point system is the following. A major obstacle in Dominion tournaments is lacking enough games to distinguish which player is best. So you want to get at much information from each game as possible.

If you only know the winner of a game, there are only 4 possible outcomes. If you know the finish order (1st-4th) of the players, there are 4! = 24 possible outcomes. So, 2 bits of information from winner-takes-all, and ~4.6 bits of information if you know the finish order, which is more than twice as much.

From a statistical perspective, I totally agree with you.

But I think the game design trumps the statistics.  I think playing to win, all else be damned, is more in the spirit of the rules.

I love you man.

The only thing different between you and me is you put "I think" in front of everything - you're so goddamn polite.

Also, this discussion might be useful.

http://boardgamegeek.com/thread/457599/discussion-on-a-universal-tournament-format-for-do/page/1

Although it might not be, because at least no one here is proposing rewarding players for having a high score.

I don't think it helps the discussion, but it makes me look old and experienced!

[blueblimp]

One good thing about a point system is the following. A major obstacle in Dominion tournaments is lacking enough games to distinguish which player is best. So you want to get at much information from each game as possible.

If you only know the winner of a game, there are only 4 possible outcomes. If you know the finish order (1st-4th) of the players, there are 4! = 24 possible outcomes. So, 2 bits of information from winner-takes-all, and ~4.6 bits of information if you know the finish order, which is more than twice as much.

From a statistical perspective, I totally agree with you.

But I think the game design trumps the statistics.  I think playing to win, all else be damned, is more in the spirit of the rules.

IRL and online, I always aim to maximize my rank in 3/4-player, and don't care that much about whether I win. So I think this is a matter of opinion.

[metzgerism]

One good thing about a point system is the following. A major obstacle in Dominion tournaments is lacking enough games to distinguish which player is best. So you want to get at much information from each game as possible.

If you only know the winner of a game, there are only 4 possible outcomes. If you know the finish order (1st-4th) of the players, there are 4! = 24 possible outcomes. So, 2 bits of information from winner-takes-all, and ~4.6 bits of information if you know the finish order, which is more than twice as much.

From a statistical perspective, I totally agree with you.

But I think the game design trumps the statistics.  I think playing to win, all else be damned, is more in the spirit of the rules.

IRL and online, I always aim to maximize my rank in 3/4-player, and don't care that much about whether I win. So I think this is a matter of opinion.

What if a trip to Nationals was on the line?
Would you still not care about winning then? Would you still just try and maximize your rank?

Each game in this tournament has the potential to be worth a fraction of a trip to Nationals.
The conditions are significantly different.

EDIT:
http://www.boardgamers.org/yearbook10/dompge.htm
Read that

[zxcvbn2]

FWIW, last year's world championship had a point system in the prelim round, and then a knockout phase (3-player) among the top 9. This doesn't mean they'll use it this year, just wanted to throw it in there.

Source: http://www.boardgamegeek.com/thread/714428/dominion-wm-tournament
and here: http://www.riograndegames.com/uploads/FileUpload/DominionWMrules.pdf

[WanderingWinder]

The thing is, I really don't understand how this is getting such vehement responses on both sides. In no way is one system clearly better than the other. They're just two different systems. It's like, the question is: "Is the point of the game to win, or is the point of the game to finish in as good a placement as possible?" Both are reasonable arguments, and there's no way that one is just 'clearly' better than the other, or that either of them are ridiculous. Having a point structure does not make it non-competitive, it just makes it different.
If you look at something like auto-racing, they give points for how well you finish, not just where you place. But there are other sports where all that matters is who win. Off the top of my head, I can't think of any sports where there's a field of participants (i.e., it's not head-to-head), where they don't give more credit to 2nd place than 8th place. That doesn't make such a system 'wrong' or 'unreasonable' though. Which is better than the other, is just based on what people think it should be. That's really what matters here. Which prima facie makes nobody's view ridiculous, though you can of course prefer to NOT have that.

There are a couple things to note:

1) Auto Racing doesn't have traditional playoffs.
2) All sports with traditional playoffs are 1v1 affairs.

These are HUGE differences! If you were to have an entire NASCAR season of 43-car races, then at the end say "we're going to have a 16-car bracket where it's just you against one other driver," fans would revolt!

The thing with dominion is that this makes more significant gameplay changes, probably, than you'd see in other things. Particularly in the endgame. Again, it's not that one way is right or wrong. How can you say that playing for second is wrong? How can you say that going all out for a win is wrong? Strategically, it's all down to how the tournament is set up. Which isn't clear one way or the other.

I am under the assumption that there will be a decisive bracket at Nationals, as well as one in this qualifier. THIS is what drives my argument against a point system. If there's no point system when it's most important, you shouldn't have one at any other time. Changing the parameters of "success" mid-tournament IS metagaming, no matter how many leagues do it. There are many sports leagues that DO fail at this, however (the NHL being an aggravating example) - sometimes it isn't the teams who win the most games that get invited to the tournament.

And that IS wrong.

I personally advocate for a points system, but one which has disproportionate favoring to the winner. Because I think that playing for a win is an important aspect of the game, and I think you should take the risks, particularly strategically. At the same time, I feel like 2nd and 4th shouldn't get treated the same. There's also then the issue of weird endgame behaviour/collusion-y stuff. Having winner-take-all encourages this more than a point system, because where this really comes up is in situations where it's not possible for a player to win - then they can do anything to throw the game to one guy or another, with no negative repurcussions for themselves. If 3rd is better than 4th, then this would only come into play when 4th place can't possibly even get 3rd, which is going to happen a lot less than 4th being unable to grab 1st. So this is another reason I prefer some point system.

The 'it's harder to set up' thing seems to be a weak argument to me, because it's not much harder to set up, and, more importantly, the people who would be doing the setting up seem willing to do this little extra.

So I would advocate something like 6/3/1/0. There's also the issue of the odd 3-player games, which I would probably have as - everyone plays four player games, the odd people out get buys, or probably preferentially, quasi-byes, i.e. 1 person odd gets a full bye, 2 people odd play a 2-player, where (if we assume a 6/3/1/0 point system for the moment), 1st gets 6 and 2nd gets 3, 3 odd people play a 3-player where 1st gets 6, 2nd gets 3, 3rd gets 1. But it's a bit of a thorny issue - someone is getting the better end of a luck stick, no matter how you slice it.
But more than this, I advocating trying to set up whatever point system to match what will actually be at nationals as closely as possible.

Agreed with the bold, far more than any other thing that's been said in this thread.

I disagree. I actually think NASCAR fans would love that. Also, the NBA and particularly NHL specifically do change their formats for the playoffs. In principle, I don't think you should change systems partway through, unless there is some compelling reason to do so, but it's certainly not crazy or clearly WRONG. It's just something different. I find it incredibly amusing that you just label so may people whose opinions are different from yours as flatly wrong. Kind of an arrogant-looking move. I think the big reason you're getting such a bristled response here isn't so much what you're saying, but how you're saying it. It's like you're being abrasive for abrasiveness's sake, or at least that's how it's coming off.
I presumed that the semis and finals would have the same scoring system as you have in the previous stages. This isn't an argument for no point system. It's an argument for the same method throughout the tournament. But what that method is, is what we're debating.
I certainly don't have a problem with meta-gaming per se. Indeed, I tend to like it. Indeed, there are whole games where metagaming is the hugest part of the game. Dominion is not and never will be one, but I don't see why it can't be a part.

[WanderingWinder]

One good thing about a point system is the following. A major obstacle in Dominion tournaments is lacking enough games to distinguish which player is best. So you want to get at much information from each game as possible.

If you only know the winner of a game, there are only 4 possible outcomes. If you know the finish order (1st-4th) of the players, there are 4! = 24 possible outcomes. So, 2 bits of information from winner-takes-all, and ~4.6 bits of information if you know the finish order, which is more than twice as much.

From a statistical perspective, I totally agree with you.

But I think the game design trumps the statistics.  I think playing to win, all else be damned, is more in the spirit of the rules.

IRL and online, I always aim to maximize my rank in 3/4-player, and don't care that much about whether I win. So I think this is a matter of opinion.

What if a trip to Nationals was on the line?
Would you still not care about winning then? Would you still just try and maximize your rank?

Each game in this tournament has the potential to be worth a fraction of a trip to Nationals.
The conditions are significantly different.

EDIT:
http://www.boardgamers.org/yearbook10/dompge.htm
Read that

That a trip to nationals is on the line is totally irrelevant. Why would that make a difference?

[metzgerism]

The thing is, I really don't understand how this is getting such vehement responses on both sides. In no way is one system clearly better than the other. They're just two different systems. It's like, the question is: "Is the point of the game to win, or is the point of the game to finish in as good a placement as possible?" Both are reasonable arguments, and there's no way that one is just 'clearly' better than the other, or that either of them are ridiculous. Having a point structure does not make it non-competitive, it just makes it different.
If you look at something like auto-racing, they give points for how well you finish, not just where you place. But there are other sports where all that matters is who win. Off the top of my head, I can't think of any sports where there's a field of participants (i.e., it's not head-to-head), where they don't give more credit to 2nd place than 8th place. That doesn't make such a system 'wrong' or 'unreasonable' though. Which is better than the other, is just based on what people think it should be. That's really what matters here. Which prima facie makes nobody's view ridiculous, though you can of course prefer to NOT have that.

There are a couple things to note:

1) Auto Racing doesn't have traditional playoffs.
2) All sports with traditional playoffs are 1v1 affairs.

These are HUGE differences! If you were to have an entire NASCAR season of 43-car races, then at the end say "we're going to have a 16-car bracket where it's just you against one other driver," fans would revolt!

The thing with dominion is that this makes more significant gameplay changes, probably, than you'd see in other things. Particularly in the endgame. Again, it's not that one way is right or wrong. How can you say that playing for second is wrong? How can you say that going all out for a win is wrong? Strategically, it's all down to how the tournament is set up. Which isn't clear one way or the other.

I am under the assumption that there will be a decisive bracket at Nationals, as well as one in this qualifier. THIS is what drives my argument against a point system. If there's no point system when it's most important, you shouldn't have one at any other time. Changing the parameters of "success" mid-tournament IS metagaming, no matter how many leagues do it. There are many sports leagues that DO fail at this, however (the NHL being an aggravating example) - sometimes it isn't the teams who win the most games that get invited to the tournament.

And that IS wrong.

I personally advocate for a points system, but one which has disproportionate favoring to the winner. Because I think that playing for a win is an important aspect of the game, and I think you should take the risks, particularly strategically. At the same time, I feel like 2nd and 4th shouldn't get treated the same. There's also then the issue of weird endgame behaviour/collusion-y stuff. Having winner-take-all encourages this more than a point system, because where this really comes up is in situations where it's not possible for a player to win - then they can do anything to throw the game to one guy or another, with no negative repurcussions for themselves. If 3rd is better than 4th, then this would only come into play when 4th place can't possibly even get 3rd, which is going to happen a lot less than 4th being unable to grab 1st. So this is another reason I prefer some point system.

The 'it's harder to set up' thing seems to be a weak argument to me, because it's not much harder to set up, and, more importantly, the people who would be doing the setting up seem willing to do this little extra.

So I would advocate something like 6/3/1/0. There's also the issue of the odd 3-player games, which I would probably have as - everyone plays four player games, the odd people out get buys, or probably preferentially, quasi-byes, i.e. 1 person odd gets a full bye, 2 people odd play a 2-player, where (if we assume a 6/3/1/0 point system for the moment), 1st gets 6 and 2nd gets 3, 3 odd people play a 3-player where 1st gets 6, 2nd gets 3, 3rd gets 1. But it's a bit of a thorny issue - someone is getting the better end of a luck stick, no matter how you slice it.
But more than this, I advocating trying to set up whatever point system to match what will actually be at nationals as closely as possible.

Agreed with the bold, far more than any other thing that's been said in this thread.

I disagree. I actually think NASCAR fans would love that. Also, the NBA and particularly NHL specifically do change their formats for the playoffs. In principle, I don't think you should change systems partway through, unless there is some compelling reason to do so, but it's certainly not crazy or clearly WRONG. It's just something different. I find it incredibly amusing that you just label so may people whose opinions are different from yours as flatly wrong. Kind of an arrogant-looking move. I think the big reason you're getting such a bristled response here isn't so much what you're saying, but how you're saying it. It's like you're being abrasive for abrasiveness's sake, or at least that's how it's coming off.
I presumed that the semis and finals would have the same scoring system as you have in the previous stages. This isn't an argument for no point system. It's an argument for the same method throughout the tournament. But what that method is, is what we're debating.
I certainly don't have a problem with meta-gaming per se. Indeed, I tend to like it. Indeed, there are whole games where metagaming is the hugest part of the game. Dominion is not and never will be one, but I don't see why it can't be a part.

Again, I don't intend to be abrasive or offensive, but I flatly disagree with having a point system and I have been running tournaments for years...I'd be shocked if anyone on DS forums has more experience than I do in this regard. I apologize for being arrogant.

I also disagree with you about metagaming being an acceptable part of the gameplay. There's only so much you can control as a tournament organizer, but you don't have to make it worse (which is what I believe a point system would encourage). This is a Dominion tournament, not a Dominion-plus-politics tournament.

---

One big thing I'm looking at here is the relation to 2p Dominion. In two player, a point system as recommended is non-existent, and you play to win (because 2nd place is losing). Why should that spirit change just because we have more players?

[WanderingWinder]

The thing is, I really don't understand how this is getting such vehement responses on both sides. In no way is one system clearly better than the other. They're just two different systems. It's like, the question is: "Is the point of the game to win, or is the point of the game to finish in as good a placement as possible?" Both are reasonable arguments, and there's no way that one is just 'clearly' better than the other, or that either of them are ridiculous. Having a point structure does not make it non-competitive, it just makes it different.
If you look at something like auto-racing, they give points for how well you finish, not just where you place. But there are other sports where all that matters is who win. Off the top of my head, I can't think of any sports where there's a field of participants (i.e., it's not head-to-head), where they don't give more credit to 2nd place than 8th place. That doesn't make such a system 'wrong' or 'unreasonable' though. Which is better than the other, is just based on what people think it should be. That's really what matters here. Which prima facie makes nobody's view ridiculous, though you can of course prefer to NOT have that.

There are a couple things to note:

1) Auto Racing doesn't have traditional playoffs.
2) All sports with traditional playoffs are 1v1 affairs.

These are HUGE differences! If you were to have an entire NASCAR season of 43-car races, then at the end say "we're going to have a 16-car bracket where it's just you against one other driver," fans would revolt!

The thing with dominion is that this makes more significant gameplay changes, probably, than you'd see in other things. Particularly in the endgame. Again, it's not that one way is right or wrong. How can you say that playing for second is wrong? How can you say that going all out for a win is wrong? Strategically, it's all down to how the tournament is set up. Which isn't clear one way or the other.

I am under the assumption that there will be a decisive bracket at Nationals, as well as one in this qualifier. THIS is what drives my argument against a point system. If there's no point system when it's most important, you shouldn't have one at any other time. Changing the parameters of "success" mid-tournament IS metagaming, no matter how many leagues do it. There are many sports leagues that DO fail at this, however (the NHL being an aggravating example) - sometimes it isn't the teams who win the most games that get invited to the tournament.

And that IS wrong.

I personally advocate for a points system, but one which has disproportionate favoring to the winner. Because I think that playing for a win is an important aspect of the game, and I think you should take the risks, particularly strategically. At the same time, I feel like 2nd and 4th shouldn't get treated the same. There's also then the issue of weird endgame behaviour/collusion-y stuff. Having winner-take-all encourages this more than a point system, because where this really comes up is in situations where it's not possible for a player to win - then they can do anything to throw the game to one guy or another, with no negative repurcussions for themselves. If 3rd is better than 4th, then this would only come into play when 4th place can't possibly even get 3rd, which is going to happen a lot less than 4th being unable to grab 1st. So this is another reason I prefer some point system.

The 'it's harder to set up' thing seems to be a weak argument to me, because it's not much harder to set up, and, more importantly, the people who would be doing the setting up seem willing to do this little extra.

So I would advocate something like 6/3/1/0. There's also the issue of the odd 3-player games, which I would probably have as - everyone plays four player games, the odd people out get buys, or probably preferentially, quasi-byes, i.e. 1 person odd gets a full bye, 2 people odd play a 2-player, where (if we assume a 6/3/1/0 point system for the moment), 1st gets 6 and 2nd gets 3, 3 odd people play a 3-player where 1st gets 6, 2nd gets 3, 3rd gets 1. But it's a bit of a thorny issue - someone is getting the better end of a luck stick, no matter how you slice it.
But more than this, I advocating trying to set up whatever point system to match what will actually be at nationals as closely as possible.

Agreed with the bold, far more than any other thing that's been said in this thread.

I disagree. I actually think NASCAR fans would love that. Also, the NBA and particularly NHL specifically do change their formats for the playoffs. In principle, I don't think you should change systems partway through, unless there is some compelling reason to do so, but it's certainly not crazy or clearly WRONG. It's just something different. I find it incredibly amusing that you just label so may people whose opinions are different from yours as flatly wrong. Kind of an arrogant-looking move. I think the big reason you're getting such a bristled response here isn't so much what you're saying, but how you're saying it. It's like you're being abrasive for abrasiveness's sake, or at least that's how it's coming off.
I presumed that the semis and finals would have the same scoring system as you have in the previous stages. This isn't an argument for no point system. It's an argument for the same method throughout the tournament. But what that method is, is what we're debating.
I certainly don't have a problem with meta-gaming per se. Indeed, I tend to like it. Indeed, there are whole games where metagaming is the hugest part of the game. Dominion is not and never will be one, but I don't see why it can't be a part.

Again, I don't intend to be abrasive or offensive, but I flatly disagree with having a point system and I have been running tournaments for years...I'd be shocked if anyone on DS forums has more experience than I do in this regard. I apologize for being arrogant.

I also disagree with you about metagaming being an acceptable part of the gameplay. There's only so much you can control as a tournament organizer, but you don't have to make it worse (which is what I believe a point system would encourage). This is a Dominion tournament, not a Dominion-plus-politics tournament.

---

One big thing I'm looking at here is the relation to 2p Dominion. In two player, a point system as recommended is non-existent, and you play to win (because 2nd place is losing). Why should that spirit change just because we have more players?

If you don't want to be abrasive, try not cursing, not telling people that their OPINIONS are WRONG in all caps (there's a big difference between 'I disagree' and 'You're WRONG'), not trying to argue via reductio ad absurdum, etc. People don't like this.

This (the analogy to 2p as you make it) is an arbitrary way of looking at it. I can just as easily make the argument that 2p dominion is all about not getting last, and why should that change by adding another player.
But also, why should 2p, 3p, and 4p be anything like the same game? They're totally different in almost every respect.

I seriously doubt you have the most experience with tournaments of anyone on this forum. Be shocked. You've barely been here, so I wouldn't expect you to know, but... well, I'm also not sure how this is relevant. (Again, this comes off as really arrogant.)
The bigger point is that you're rather arbitrarily deciding that preferring 2nd to 4th is not part of dominion. You have said this many times, but the only evidence you give is that it's not. Which is no evidence at all. You say that having a point system would make the tournament worse, but again, you give no reason. People are giving reasons on the other side. You're just presenting your summary opinion as fact.
The game is what you want it to be. It's a game.

[blueblimp]

One big thing I'm looking at here is the relation to 2p Dominion. In two player, a point system as recommended is non-existent, and you play to win (because 2nd place is losing). Why should that spirit change just because we have more players?

I think of 2p in points too. A win is 2 points, tie 1 point, loss 0 points. So if I'm in a situation where I feel I have less than a half chance to win, then I'm pretty happy to get a tie.

Of course if there is a tournament meaning assigned to rank, then I would play according to that. I'm just saying that in casual play, I can be fairly happy with 2nd in a 4-player game, much more than with 4th, and I'll happily end the game in 2nd if it prevents me from dropping lower.

[metzgerism]

If you don't want to be abrasive, try not cursing, not telling people that their OPINIONS are WRONG in all caps (there's a big difference between 'I disagree' and 'You're WRONG'), not trying to argue via reductio ad absurdum, etc. People don't like this.

I put words in caps or italics that I mean to stress. I don't believe I said anyone was flat out wrong, anywhere, especially not in caps...?

This (the analogy to 2p as you make it) is an arbitrary way of looking at it. I can just as easily make the argument that 2p dominion is all about not getting last, and why should that change by adding another player.
But also, why should 2p, 3p, and 4p be anything like the same game? They're totally different in almost every respect.

How is it arbitrary? I didn't pick "2 player" out of a hat - over 99% of my games are played 2-player, and I'm sure the ratio is similar for most people on this board, including you.

That said, you're absolutely right that they are not the same game (and shouldn't be held to the same standard).

I seriously doubt you have the most experience with tournaments of anyone on this forum. Be shocked. You've barely been here, so I wouldn't expect you to know, but...

I've run a plurality of the total online english-language Dominion tournaments/competitions. There have been 33 that I know of. I have operated 13 and was founder of the league that ran another 12. All of those came after years of running other competitions on other boards for other games.

You don't know me very well, but this is my bread & butter.

[metzgerism]

This thread is off-track, so I wanted to repeat my latest recommendation for a format:

- All players organized into ~12-player pools. At the end of the pool phase, the top 1-2 players in each pool get a bracket bye, and wild cards go to play-in games.
- You may play 3p or 4p games against opponents in your pool only. You may play any single opponent repeatedly UNTIL you win a game they lost, or vice versa - effectively 2nd, 3rd, and 4th place are all "ties," and ties should be replayed.
- Because of the open nature of the pools, not every matchup will be represented by a game and there will likely be some holes in the results. Being active early and often will help you find your opponents before you can't organize a game.
- Players are ranked on "wins," which are actually the number of other players beaten. A regular 4p win is worth 3 "wins," a 3p win is worth 2. Shared victory is a tie with the other victorious players. Tiebreakers are given for getting 2nd place in a matchup you lost.
- Not sure how the bracket would work, only that there should be 3p and 4p games if there are 3p and 4p games in the pool phase.

[O]

A question completely irrelevant to the above conversation: can minors sign up? 

[Kirian]

The real argument now is whether winner-take-all or a points-based system is better for the heats.  And to those who support the former--metzgerism and rrenaud chief among them, I think--I have a simple question:

Name a sport, professional or Olympic, in which more than two players compete in each round, but only one player from each match in each round advances.

I can't find one, but maybe I'm wrong.  Golf is the only commonly-followed sport* I can think of that are not one-on-one games, and it uses a point system.  Swimming and track sports use heats from which multiple athletes advance from each race.

*NASCAR was mentioned, but, sorry, just no.

[O]

Equally hard question: Name a professional/olympic sport that isn't golf and has 3+ teams competing in the same game/match.

Archery?

[metzgerism]

The real argument now is whether winner-take-all or a points-based system is better for the heats.  And to those who support the former--metzgerism and rrenaud chief among them, I think--I have a simple question:

Name a sport, professional or Olympic, in which more than two players compete in each round, but only one player from each match in each round advances.

I can't find one, but maybe I'm wrong.  Golf is the only commonly-followed sport* I can think of that are not one-on-one games, and it uses a point system.  Swimming and track sports use heats from which multiple athletes advance from each race.

*NASCAR was mentioned, but, sorry, just no.

I can't think of any. There are few confrontational sports (if any) that are not 1v1, and therefore a heat-based system is more preferable. There have been competitions in the past where a single qualifier is returned from a grouping.

I'm also not necessarily advocating that single qualifier thing, so I'm not sure where this discussion is heading...

[metzgerism]

Equally hard question: Name a professional/olympic sport that isn't golf and has 3+ teams competing in the same game/match.

Archery?

Races, like cycling and motorsports. Those usually don't have a winner-take-all scenario for each individual race.

[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.

[theory]

I am more interested in asking -- how long is the slowest 4p match likely to take?  Is 30min a reasonable upper limit?  I think it is, but all the 4p games I've played in have me in them and are therefore subject to sampling bias.

[Captain_Frisk]

I am more interested in asking -- how long is the slowest 4p match likely to take?  Is 30min a reasonable upper limit?  I think it is, but all the 4p games I've played in have me in them and are therefore subject to sampling bias.

I bet the longest will be an hour.

All you need is kc, pool, young witch, secret chamber bane, minion.  I've played 30 min 2p matches before.   

[WanderingWinder]

If you don't want to be abrasive, try not cursing, not telling people that their OPINIONS are WRONG in all caps (there's a big difference between 'I disagree' and 'You're WRONG'), not trying to argue via reductio ad absurdum, etc. People don't like this.

I put words in caps or italics that I mean to stress. I don't believe I said anyone was flat out wrong, anywhere, especially not in caps...?

Addressing this privately, as flame wars are baddddd, mmkay?

This (the analogy to 2p as you make it) is an arbitrary way of looking at it. I can just as easily make the argument that 2p dominion is all about not getting last, and why should that change by adding another player.
But also, why should 2p, 3p, and 4p be anything like the same game? They're totally different in almost every respect.

How is it arbitrary? I didn't pick "2 player" out of a hat - over 99% of my games are played 2-player, and I'm sure the ratio is similar for most people on this board, including you.

That said, you're absolutely right that they are not the same game (and shouldn't be held to the same standard).

You're missing my point. I'm not saying using 2p is arbitrary. I'm saying that viewing 2p as 'the whole point is to win' is arbitrary. You can just as easily view it as 'the whole point is not to lose'. Or 'the whole point is to not get last' (which is the most supported by the rulebook, IMO, ('...rejoice in their shared victory...')). Or 'the whole point is to maximize your position'. Which is what the point system advocates, basically. You continue to privilege your own position, without considering the other side.

I seriously doubt you have the most experience with tournaments of anyone on this forum. Be shocked. You've barely been here, so I wouldn't expect you to know, but...

I've run a plurality of the total online english-language Dominion tournaments/competitions. There have been 33 that I know of. I have operated 13 and was founder of the league that ran another 12. All of those came after years of running other competitions on other boards for other games.

You don't know me very well, but this is my bread & butter.

I don't know you very well, because you only just showed up here! And now you're trying to act like THE authority. Which is a frustrating thing. It's your 'bread-and-butter'? You do this for a living? Certainly not the dominion thing. 15 tournaments online over the past several years is pretty unimpressive, considering the size of things. Trust me, I know many people who have WAY more organizing experience than this (not with Dominion itself, but more established competitive games), including one or two who are on the site (not that they've posted in this thread). I don't know what their positions here would be, because they don't deal with multiplayer messes really. But suffice it to say, unless you've really buried the lead in those 'years of running competitions'. Finally, if you're counting BGGDL as 25 separate things, a) don't take credit for the 12 you didn't run, but more important, b) there have been WAY more than 33.

[WanderingWinder]

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.

I think you're underestimating how big an effect this has on how multiplayer is played.

[WanderingWinder]

I am more interested in asking -- how long is the slowest 4p match likely to take?  Is 30min a reasonable upper limit?  I think it is, but all the 4p games I've played in have me in them and are therefore subject to sampling bias.

I bet the longest will be an hour.

All you need is kc, pool, young witch, secret chamber bane, minion.  I've played 30 min 2p matches before.   

I would say that 90+% of games will be done in 30 minutes or less. I would guess that of the remaining 10%, 90% of those will be done within 15 minutes after. Then you get that terrible game where 4 people have reactions and everyone has attacks, and people step away from their computer and.... ick. But I don't think 30 minutes is a bad rule of thumb for round time, even with attacks.

[rrenaud]

I am more interested in asking -- how long is the slowest 4p match likely to take?  Is 30min a reasonable upper limit?  I think it is, but all the 4p games I've played in have me in them and are therefore subject to sampling bias.

We can fix a set apriori (and not tell anyone until the tournament starts), which will get rid of some variance in the game length. 

[WanderingWinder]

Fixed sets is something to bring up, because I'm pretty sure that's how it will be at nationals as well.

[metzgerism]

Fixed sets is something to bring up, because I'm pretty sure that's how it will be at nationals as well.

Unfortunately, this is true. A lot of tournaments have fixed sets for all games in a round - as long as we have distinctly segregated rounds, we might want to have players get the same array of kingdom cards throughout the tournament.

Not that I advocate for it, but it's probably how the nationals are going to be.

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.

If that is the case, where more wins invariably equals a higher ranking, then I'm not worried about it at all - furthermore, that means a traditional point system would be merely cosmetic.

I am more interested in asking -- how long is the slowest 4p match likely to take?  Is 30min a reasonable upper limit?  I think it is, but all the 4p games I've played in have me in them and are therefore subject to sampling bias.

I bet the longest will be an hour.

All you need is kc, pool, young witch, secret chamber bane, minion.  I've played 30 min 2p matches before.   

I second this.

[theory]

A question completely irrelevant to the above conversation: can minors sign up? 

Yes.  If you win you must be accompanied by a parent who will not have their expenses paid for.