Outwitters WORLD CUP Winner - Alvendor (SWEDEN)

[Ravernoth]

Cool, thanks Combatex. If you or awpertunity end up doing the groups again, my friend samura| has been promoted, so would jump up the seedings:

1 - samura|
2 - UK
3 - Gifted, 988, 64

Thanks awpertunity for setting this up!

[Judment]

I would also support a full rules update

[awpertunity]

(09-04-2012 03:28 AM)CombatEX Wrote:  

(09-03-2012 08:46 PM)Ravernoth Wrote:  

(09-03-2012 02:56 PM)CombatEX Wrote:  Also here are the groups using the adjusted seeding. I arranged the groups the way you suggested (1-9-17-25, 2-10-18-26, 3-11-19-27, etc).

Actually, this is a bit off - this guarantees that higher seeds are placed together:

1 > 2
9 > 10
17 > 18
25 > 26

Shouldnt seeds be placed in 'pots' and randomly drawn into each group like in the World Cup?
i.e.

Pot 1: 1-8
Pot 2: 9-16
Pot 3: 17-24
Pot 4: 25-32

And an ideal setup may be to have Pot 1 not random - the seeds placed in the groups to have the expected seeded matchups in the round of 8, i.e:

Left bracket:
A - 1st seed
C - 8th seed

E - 4th seed
G - 5th seed

Right bracket:
B - 2nd seed
D - 7th seed

F - 3rd seed
H - 6th seed

You have a good point there. Ultimately awpertunity is the one in charge of this tournament so I defaulted to what he mentioned in an earlier post. I'll do a random draw now though with top 8 seeds arranged as you suggested though to see how it would turn out. Of course none of this matters unless awpertunity agrees =)


Adjustments
- Distributed top 8 seeds into respective groups as suggested by Ravernoth
- Randomly assigned each group with a player from 9-16, 17-24, and 25-32 (I used random.org)

Revised Groups
http://img138.imageshack.us/img138/4461/...groups.png

The country tags correspond to the respective positions of the players in the groups.

Here is a rough estimate of group strengths (Strongest -> Weakest)

Group H (+6)
Group B (+5)
Group E (+3)
Group F (+1)
Group C (+0)
Group A (-1)
Group G (-3)
Group D (-11)

Note: In reality there are probably many players who are underestimated by their leagues (80%+ w/l without promotion) not to mention points/win has flaws as a means of estimating skill. In other words, don't take this too seriously.

Thank you both for your inputs and help!

I have finished creating the groups in the following manner:

The top 16 seeds were placed such that the expected matches (according to the seeds) will be in the standard tournament format, i.e., 1 vs 16, 2 vs 15, and so on.

Quarterfinals will be 1 vs 8, 2 vs 7, and so on.
Semifinals are 1 vs 4, 2 vs 3.
Finals are 1 vs 2.

Now let's see some upsets!

The remaining 16 players were placed into two pots as suggested for seeds 17-24 and 25-32 and one player from each pot has been randomly assigned to a group.


RULES HAVE BEEN UPDATED ON ORIGINAL POST. Please read and be familiar with all the rules. The tournament has not started yet, so although I will post the final Groups shortly, please do not start any matches yet.


As always: questions, comments, criticisms and suggestions are appreciated!

---

PDF file has been updated. Everyone please make sure I have your GameCenter Nickname written correctly. (Are spaces really in your names '27 Pastore' and 'Da Wizerd'?)


Combat: In case you were wondering I used your adjusted seeding for group placements, with the exception of putting the two players without wins posted into the higher pot just because they have so many more points than the others in Gifted.


To the players who have been posting updated rankings. Thanks for your efforts/honesty but I have simply calculated your seed based on the time of your original post. Rankings change a lot!

[CombatEX]

(09-05-2012 10:33 AM)awpertunity Wrote:  Combat: In case you were wondering I used your adjusted seeding for group placements, with the exception of putting the two players without wins posted into the higher pot just because they have so many more points than the others in Gifted.

Sounds good. Glad I could be of some help =)

A couple things though:

1. In your pdf should samura- be samura|?

2. Would you be willing to organize replays somewhere for the tournament like GreatGonzales did for the first community tournament? I'd like to watch the games =) I realize that it is already a requirement for people to upload the replays to this thread, but it would be nice to have them organized for easy viewing.

Anyway, good luck everyone in the tournament!

[awpertunity]

(09-05-2012 11:11 AM)CombatEX Wrote:  1. In your pdf should samura- be samura|?

Fixed, thanks.

(09-05-2012 11:11 AM)CombatEX Wrote:  2. Would you be willing to organize replays somewhere for the tournament like GreatGonzales did for the first community tournament? I'd like to watch the games =) I realize that it is already a requirement for people to upload the replays to this thread, but it would be nice to have them organized for easy viewing.

Absolutely!

[GreatGonzales]

Hey Awpertunity, I know this is late but I just had a thought about the rules, specifically the way you are assigning points. I am imagining circumstances where someone wins with a number of turns that are just beyond the better point bracket. Using these guidelines:

The PF awarded will be as follows:
SMALL MAPS (1 wit space each)
+4 points for 12 turns or less
+3 points for 28 turns or less
+2 points for 48 turns or less
+1 point for 49 turns or more

LARGE MAPS (2 wit spaces each)
+4 points for 14 turns or less
+3 points for 32 turns or less
+2 points for 56 turns or less
+1 point for 57 turns or more

So for instance, on a small map, someone who wins on turn 30 for both games would be getting the same number of points as someone who wins on turn 47. That's not really fair, and I think we can do better than this by using a formula to calculate the number of points - this would mean decimals - according to the exact number of turns.

So, for small maps you would start with 4 points and subtract points for each turn after 12 using this formula:

Points (small)= 4 - ((Total number of turns - 12) * .05)

For large maps, we would change the coefficient:

Points (large) = 4 - ((Total number of turns - 14) * .04)

Some examples:

Winning on Glitch on turn 30 would give you 3.10 points.
Winning on Glitch on turn 47 would give you 2.25 points.

Winning on Long Nine on turn 18 would give you 3.84 points.
Winning on Sweetie Plains on turn 44 would give you 2.80 points.

You can play with the coefficient if you want to tweak the influence of number of turns up or down. And I would probably recommend implementing artificial ceiling and floors of 4 and 1 points. But I am in favor of this sort of system because it will more precisely capture skill level, and ties are going to be very unlikely.

What do you think?

---

It might be better to make the coefficients a bit more harsh. Currently, in order to score 1 point for a win on a small map you need to take 72 turns; on a large map you would need to take 89 turns. That's a bit high.. Perhaps you should use:

Points (small)= 4 - ((Total number of turns - 12) * .06)

Points (large) = 4 - ((Total number of turns - 14) * .05)

With these, to score 1 point on a small map you would need 62 turns; to score 1 point on a large map you would need 74 turns.

Or, you know, something. It doesn't matter all that much, really.

---

Here are those examples with the tweaked formulas, I think this is a little better:

Winning on Glitch on turn 30 would give you 2.92 points.
Winning on Glitch on turn 47 would give you 1.90 points.

Winning on Long Nine on turn 18 would give you 3.80 points.
Winning on Sweetie Plains on turn 44 would give you 2.50 points.

[Dawizerd]

Yes, my game center ID is

Da Wizerd


Thanks

[Ravernoth]

awpertunity - it all looks good - just a thought, the group tables might be easier to read without the colours. Or could have the rows white and light grey.

Greatgonzales - The points are sufficient for the group stage since draws are allowed, and in the case of ties, the head to head number of turns is a tiebreaker. In the knockout stage, in the case of a draw, the player who won in fewer turns wins.

[Magnum]

Yes awpertunity, I'm sure there is a space in my game center's name.

[GreatGonzales]

(09-05-2012 05:45 PM)Ravernoth Wrote:  Greatgonzales - The points are sufficient for the group stage since draws are allowed, and in the case of ties, the head to head number of turns is a tiebreaker. In the knockout stage, in the case of a draw, the player who won in fewer turns wins.

Oh ok, then I guess my approach isn't necessary. I'm not totally up to speed on the rules for this tournament yet - so points will be assigned but if there is ever a tie then the tie will be resolved with the total number of turns taken across all games? Or just the wins? Or, will ties only be resolved between a head to head matchup, not across all matches in a player pool? Naturally, then, awpertunity will have to be capturing the precise number of turns taken in every recorded game.