|
Post by Kayz on Mar 29, 2012 19:48:58 GMT
Of course first player ratings, that was the main idea from the beginning.
And we don't draw 1 in one branch, 2 in the other and 3 in one again, don't do 1357 and 2468, this is not fair, only 1458 and 2367 is fair.
|
|
|
Post by khamski on Mar 29, 2012 20:08:20 GMT
Seems reasonable
|
|
|
Post by Zolo-du on Mar 29, 2012 20:23:20 GMT
What's wrong with top 1inA, 3inB, 5inC, 7inD, 2inE, 4inF, 6inG, 8inH ?? A problem lol. Promotion obviously first and second then 1Avs2H 1Bvs2G 3Cvs2F 1Dvs2E and 1Evs2A 1Fvs2B 1Gvs2C 1Hvs2D ...its normal rule.
|
|
|
Post by Zolo-du on Mar 29, 2012 20:29:33 GMT
1458 and 2367 unfair becouse 1v8 = 2v7?? lol if 13572468 then 1v7 =2v8 it's ok
|
|
|
Post by khamski on Mar 29, 2012 21:05:27 GMT
Zolo: Back to roots? )) lol why not? Why not coming back to roots?
Groups are made using subjective rating. After that everything is fixed like in previous Crespo driven editions.
|
|
|
Post by Kayz on Mar 29, 2012 21:13:58 GMT
I disagree Zolo, check the onl playoff, it's the fairest.
1v8 and 2v7, both together added 9, the first ranking has the privilege to get a "weaker" player than the 2nd, it's fair indeed.
|
|
|
Post by Zolo-du on Mar 29, 2012 22:00:58 GMT
lol no " + " only "-" 1v8 =large difference in the level 2v7=smaller difference in the level ,understand now?
btw.this method first and second players only meet in the finals ..
But for me it does not matter.
|
|
|
Post by Kayz on Mar 30, 2012 5:14:16 GMT
Well how you want to make the same difference for all levels if you start with 1v7 and 2v8? 3v9? :/ we only got 8 first in PO stage, only option would be 1v5, 2v6, 3v7, 4v8 and then you tell me why the 8th player deserved to play only the 4th ranker, though the 5th played much better? (in this case has a higher rank) Though I agree, that it would be the best, to have also the same distances between ranks, but that's sadly not possible, if we use the 1st vs 2nd of group stage rule. Anyway thansk a lot for writing here with ideas Zolo Edit: What's wrong with top 1inA, 3inB, 5inC, 7inD, 2inE, 4inF, 6inG, 8inH ?? A problem lol. Promotion obviously first and second then 1Avs2H 1Bvs2G 3Cvs2F 1Dvs2E and 1Evs2A 1Fvs2B 1Gvs2C 1Hvs2D ...its normal rule. To this point there is the problem: We 3 agreed that we want random branches, for us it's no problem if for example ranks 2 and 3 or 1 and 4 meet already in the 2nd stage of the tree, now Zemke and me only want to separate 1458 and 2367 on the two branches, and then let the rest be random, so we don't have the top 4 clatching in 1st stage, and 1, 4 and 2, 3 yet in final. (If they come through)
|
|
|
Post by khamski on Mar 30, 2012 10:16:42 GMT
Okay. Lets make it more visual once again.
Group games finished. We have 8 first placers and 8 second placers.
We make a draw with one restriction: "Firsts meet Seconds"
Done. We have pairs.
Now we need to put them on the tree in a fair manner.
We rate pairs according to a subjective rank of the 1st placed player in each pair.
Ok. Got it.
Branch One We put pairs1,4,5,8 in one pot. Shuffle. Take out one pair. Take out other pair.
Okay we have 2 games of the tree filled
Take out one pair. Take out other pair. Now we have whole branch filled. 4 games are in place.
Branch Two Same process as above.
Tree is filled.
Of course 1|4, semis and final are predefined.
Tree is ready.
Am i right?
|
|
|
Post by Kayz on Mar 30, 2012 10:22:25 GMT
That's the best visual sight of the idea exactly. Crespo's old system does almost the same thing, but you have the whole tree pre-defined, means you could manipulate the branches with playing maybe in a way you get the 2nd place for sure in the group stage. Since the branches in the idea are completely shuffled, it's completely random, but fair
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Mar 30, 2012 14:47:03 GMT
Hey around,
I vote for a compromis*. Its based on above mentioned restriction, that we have two pots of 1. and 2. placed guys from group stage.
A second restriction could be, that you don't play against your group opponent in 1/8 Final. (but this shouldn't be part of this discussion)
Third and decisive restriction:
We determine two best players of starter field. We set one at the top of the tree (branch 1) and the other at the bottom of the tree (branch 2). Now we start the draw.
on the assumption that the higher ranked player wins his games, following "extrem top" games are possible:
Quarterfinal: Player 1 vs Player 3 Semifinal: Player 1 vs Player 4 [or inverse]
This is far away of (feared) accumulation of "the" top players. But it promises still maximum of transparence and maximum of simplicity.
*This was originally proposed by Dario.
|
|
|
Post by Zolo-du on Mar 30, 2012 17:13:47 GMT
Why can not meet in the semis or in the quarterfinals first vs second??So it happens, it's normal for tournaments ...
|
|
|
Post by Kayz on Mar 30, 2012 18:19:51 GMT
So you think it's normal for tournaments, that in one branch a guy must beat for example 3 former CWT finalists to even be ABLE to get bronze, but on the other branch you just have random other people who play average to very good?
Oh man..
So of course I vote against the compromiss, I'd like to have the most fairest tourney, and that's only possible with the 1458 vs 2367 branch dividing, notice, that you still have random branches.
Worst thing that can happen: 1v4 in Quarter, 1v5 in Semi, and 1v2 in final. and for the 2nd player 2v3 in Quarter, 2v6 in Semi, and 2v1 in final.
Seems pretty equal, isn't it? Furthermore as Joschi said, the difference between 1 and 2 and the difference between 3 and 4 are not that significant, that's why you still have a very challenging tree, but reducing the treelucking factor.
|
|
|
Post by khamski on Mar 30, 2012 18:52:32 GMT
Kayz system is the fairest and simplest one. I vote for it. I visualised it for those who dont get our crappy talks. See messages above.
|
|
|
Post by Zolo-du on Mar 30, 2012 19:10:52 GMT
all good
|
|