Print

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

[DG]

Maybe a new thread could be started for round 2 results.

[Kirian]

TWO brackets. Three wasn't discussed until the 8th.

True, though it didn't look like we were actually going to hit 64+ people until 07 June.

And I at least get a pretty big tone difference. Your OP there seems more like 'oh, if we have too many, we've got to break it up, so that we can have a reasonable tournament for the number of rounds we're doing'. Which is fine. An then later it's 'oh, I think I want 3 groups for no given reason'. Which is not to say there is no reason. But you don't give one.

While I don't give a direct reason, I do reference the discussion [between Fabian and O] about the pros and cons of larger vs. smaller brackets.  Essentially, I felt Fabian's argument was stronger.

Unfortunately,  you locked the thing (to be fair, the locking is unrelated to this issue - I'm not trying to say you had any nefarious reasons for locking it or anything like that), so now us normal-folk can't quote relevant passages without jumping a bunch of hoops.

I'll go ahead and unlock it, but I'm also going to fork this thread in a few minutes here.

[Mergus]

I think it's great that Kirian organises this. I'm sure he put a lot of thought into setting this up the way it is now and I think people should wait until it's over before judging it.

For my part, I think trying this format is great and it was clear for me that people would be divided into groups if there were many signups. I don't think it's appropriate to complain about the wording that was used in the original rules posting and making such a big thing out of it. This is something that someone organises in his free time and we don't pay for it.

[popsofctown]

Segregated tournaments is just fine.  Dominion is such a broad game to master I don't expect huge skill changes overnight, so players in lower divisions don't have a chance of winning the most prestigious division anyway.

And... either the ranks matter or they don't.  If they do matter, they are a good predictor suggesting you can't actually win in the division above you.  If they don't matter, then the people you play in your division are just the same as the ones in the other divisions.  Going into my second game councilroom said that I was 0-2 against my rank 34 opponent, although I apparently have a winning record against ehunt.  Whaddayaknow.

EDIT: Ok, on double check I'm tied.  Point stands though.

[bozzball]

I have never been in a Swiss-style tournament before, but I thought it meant that even if you lose, you get to play more games. I seem not to have an opponent. Is it knock out in the early rounds?

[Fabian]

There's probably an uneven number of players in your division, meaning one of the players at 0 points got a free win ("bye") this round.

[Kirian]

I have never been in a Swiss-style tournament before, but I thought it meant that even if you lose, you get to play more games. I seem not to have an opponent. Is it knock out in the early rounds?

Your opponent in round 1 was unable to contact you for your first round match, and posted about it in the missing opponents thread (stickied).  You never posted or PMd me to claim otherwise.  By the rules this constituted a forfeit and dropout.  You did not lose your first round; you never showed up.

(Your opponent's post has since been deleted, but was up for about four days before the end of the first round.)

[Young Nick]

On another note, how do you, Kirian, decide the pairings? For the first round is it completely random? For the second, is it random among those who won and those who lost? How do you decide which competitor who won their first round match up plays against which competitor who lost theirs? I ask because I noticed that my pool (B) has 22 people, so there had to be one second round match up between someone who was 1-0 and someone who was 0-1. That happened to be me. Am I at a disadvantage because of it?

Or does challone automatically do match-ups? I'd be pretty guh if I were at a disadvantage despite being 2-0.

[Kirian]

On another note, how do you, Kirian, decide the pairings? For the first round is it completely random? For the second, is it random among those who won and those who lost? How do you decide which competitor who won their first round match up plays against which competitor who lost theirs? I ask because I noticed that my pool (B) has 22 people, so there had to be one second round match up between someone who was 1-0 and someone who was 0-1. That happened to be me. Am I at a disadvantage because of it?

Or does challone automatically do match-ups? I'd be pretty guh if I were at a disadvantage despite being 2-0.

First round was random (within each bracket).  Challonge does the calculations for the remaining rounds based on what I believe is standard, which means using seeds (even though the initial seed was randomized).  Since you're the lowest seed in the bracket (after randomization), you're "lowest" within the winning group; thus you were paired with the "highest" of the losing group (pops, in this case, who was first in the randomized seeding).

This shouldn't put you at a disadvantage; we can't predict the final score for any of your opponents right now.  In any case, pairing calculations like this tend to try to do this as few times to a single person, so you should definitely be paired with another 2-0 player next time (assuming the other five 1-0 vs. 1-0 matches don't all end in ties).

Edit:  This should be easy since there are now 2 more 2-0 players in the B bracket.

[Young Nick]

OK, thanks for clearing that up.

[jsh357]

So I have been digging through these topics and given up finding my answer.  What happens after the initial 5 Swiss rounds?  The original post says the 'top 8' move on, but how will the top 8 be determined?

[Rabid]

The top eight players after the Swiss rounds will move into a single-elimination, seeded, three-round bracket.
http://forum.dominionstrategy.com/index.php?topic=2695.0

Is this 3 separate top 8's, one for each of A B C?
Or how about top X from A, top Y from B, top Z from C to make a combined champion?

[O]

The top eight players after the Swiss rounds will move into a single-elimination, seeded, three-round bracket.
http://forum.dominionstrategy.com/index.php?topic=2695.0

Is this 3 separate top 8's, one for each of A B C?
Or how about top X from A, top Y from B, top Z from C to make a combined champion?

I think separate is more fun games and consistent with other games Swiss systems.

[Jorbles]

The top eight players after the Swiss rounds will move into a single-elimination, seeded, three-round bracket.
http://forum.dominionstrategy.com/index.php?topic=2695.0

Is this 3 separate top 8's, one for each of A B C?
Or how about top X from A, top Y from B, top Z from C to make a combined champion?

I think separate is more fun games and consistent with other games Swiss systems.

I agree. Now that we're in this mode I think it's fine to keep the brackets separate. I'm getting more into it now that the tournament rules have had time to sink in.

[Kirian]

Indeed, the top-8 eliminations will also be from each bracket.  It's consistent with finding a "champion" in each bracket.

That means that even if you've lost two matches, you still have a shot at getting into the top 8 for your bracket.  Which may or may not have been an impetus for going to smaller brackets rather than larger...

[Obi Wan Bonogi]

Now that the Tournament lobby exists is there any thought of adjusting the rules to determine who plays first in the initial game?

I think that in the top 8 giving the higher seed the home-field advantage of first player in the first game is reasonable.

[DG]

I don't see any convincing reason to use seedings or rankings for first game advantage. Is there no option to have an entirely random start player for tournament play?

[andwilk]

Since the tournament lobby didn't appear until midway through this tournament, I wouldn't recommend changing the rules now.  However, first player advantage is statistically significant and therefore should be mitigated as much as possible in future tournaments.  For future IsoDoms, I'd recommend evening out who starts as first player throughout the tournament.

[WanderingWinder]

Now that the Tournament lobby exists is there any thought of adjusting the rules to determine who plays first in the initial game?

I think that in the top 8 giving the higher seed the home-field advantage of first player in the first game is reasonable.

I don't have a dog in this fight (this particular tournament), but I don't see any reason to give higher-seeded players advantage over lower-seeded ones, except maybe that they get to play the lower seeds. Otherwise, it doesn't seem fair, really.

Unless by higher seeded, you mean they've done better so far in the tournament? Which means I'm totally mis-reading you.

[blueblimp]

Since the tournament lobby didn't appear until midway through this tournament, I wouldn't recommend changing the rules now.  However, first player advantage is statistically significant and therefore should be mitigated as much as possible in future tournaments.  For future IsoDoms, I'd recommend evening out who starts as first player throughout the tournament.

I disagree, because right now I don't think there's ANY rule stating who plays first in any game of the Bo5. I've just been assuming it's the standard (random in the first game, loser goes first in subsequent games), but it would be good at least to codify that in the rules if that's really what's supposed to happen.

Edit: And FWIW, I'd like to see first player in Game 1 stated in the bracket and then enforced by the Tournament lobby (or even starting a game and resigning), because then the logs show the rule was followed. With isotropic-decided random, there's no way to verify that the other player has actually logged in fresh. I prefer random first player rather than doing it by seeding.

[RisingJaguar]

Now that the Tournament lobby exists is there any thought of adjusting the rules to determine who plays first in the initial game?

I think that in the top 8 giving the higher seed the home-field advantage of first player in the first game is reasonable.

I would love to see some sort of home-field advantage in the future!  It'll make dominion feel like a real sport!

[Obi Wan Bonogi]

Unless by higher seeded, you mean they've done better so far in the tournament? Which means I'm totally mis-reading you.

Yes.  That is how you get seeds.  The top 8 is usually a bracket with the best record out of the swiss rounds (#1 seed) playing the #8 seed etc.  Im not sure if the semi-finals are reseeded or not? 

I just don't see the reason to use unverifiable random select for first player now that a smoother option has presented itself. 

Though I understand that it came along mid-tournament and changing the structure now could be problematic.  That is why I only suggested changes to the top 8 protocol.  Down the road I think balancing first player advantage throughout the swiss rounds would desirable as well. 

[Kirian]

Interesting questions, and I'll certainly consider them.  Certainly in a Swiss system there's something to be said for codifying first player in each round anyway (cf. black vs. white in Chess tournaments).  And while we now have the ability on Iso, whether or not we will have it on FS is another question entirely.

I don't want to change rules in the middle of the tournament, though, and while certainly the rules don't specify starting player (they probably should have done), the assumption has been that starting player is random.

Whether or not a player should be given "home" advantage is a different question that should also be considered.

[DG]

I have no objection to qualification results from the group stage defining a seeding for the remainder of the tournament with a defined first player. You do get some minor issues to resolve such as - if the number 8 seed beats the number 1 seed do they inherit 'home' advantage or will they always be 'away'?

[WanderingWinder]

Unless by higher seeded, you mean they've done better so far in the tournament? Which means I'm totally mis-reading you.

Yes.  That is how you get seeds.  The top 8 is usually a bracket with the best record out of the swiss rounds (#1 seed) playing the #8 seed etc.  Im not sure if the semi-finals are reseeded or not? 

I just don't see the reason to use unverifiable random select for first player now that a smoother option has presented itself. 

Though I understand that it came along mid-tournament and changing the structure now could be problematic.  That is why I only suggested changes to the top 8 protocol.  Down the road I think balancing first player advantage throughout the swiss rounds would desirable as well. 

You say that like it's stupidly obvious. But you can just as easily get seeds from overall ranking, which they do in all the sports I can think of that have a similar ranking system - which are only chess and tennis, of course, but there you go.