A proposal for 2023 championship

It is totally clear that the overall arena system as for qualifiers is not adequate. Many threads have been done about that.

But it is also clear that the Great and Wise Organisers chose it because it's much easier for them rather than organising matches manually like 2-4 years ago.

Also, there is a clear contradiction between the need to ensure the best players be in the finals (=goal of any selection and qualifiers everywhere) and the need to enlarge the base, to increase number of the games, to let a chance to everybody, etc.

And it's tricky. If you allow everybody, it creates a lot of lottery (unless arenas are very very long, 6-24 hours). If you don't allow everybody, it reduces the number of the players. If you let all the players together, it makes it almost impossible for the average (even rather good) player to qualify because only one is selected per arena. Etc.

In addition, as for the semifinals, to have to select only one winner per batch creates distorsions because if you have 2 very strong players together (who normally should BOTH be in the finals), you are not sure to have really the best 4 players in the final (classical issue, that can't be solved totally, but can still be diminished a lot, cf. below).

So being professionally involved in high level problem solving and used to optimisation under constraints, I had to invent a different system that would satisfy all levels' players and the admins as well, and wouldn't significantly increase their workload. Of course it's too late for 2022, but I submit you this proposal for 2023. As result of a lot of analysis of what happened in 2019, 2020, 2021 and currently in 2022.

The championship is organised in 4 stages.

STAGE 1

The organisers announce the date and the OFFICIAL BEGINNING of the championship. It's important because some players will be selected directly.

So for instance at UTC+00 on 25th of May 2023 Luke will select the 10 highest rated players (OR 9 if the incumbent Champion is not among them (whatever the reason: he was not active or is rated 14th for example...). These 10 players will not be allowed to take part in the qualifiers (if they do even inadvertently, they are DQed).

The other players, regardless of their rating (!), are allowed into arenas. The arenas are only organised during ONE week-end; there are 8 arenas with the duration of 4 hours (and 2 hours between them). I.e. Sat. and Sun. 0:00 UTC, 6:00 UTC, 12:00 UTC, 18:00 UTC. No limitation of participation of any kind. If someone is crazy enough to take part in all of them, he may.

The top 7 players in each arena are given: 60, 40, 28, 20, 14, 10, 7 points.  

The results are summarised, and then 22 players with most points are selected for the Stage 2. So someone who will have reached the 2nd place 3 times will be selected (whereas today he is not at all...). 

(Because of the duration of the arenas and the fact that 7 places give points, and because 22 players are selected overall, and because the strongest 10 players will not be able to take part, statistically all the distorsions would disappear and we shall really have the 22 next strongest players selected).

STAGE 2

The 32 players are put into two arenas A and B of 16 players only allowed (they will have to give their preferences as for schedules, and the organisers will choose the appropriate time in order to really suit all players). The duration will be 5 hours; after that the 8 first players are selected in each arena. If someone of 16 is not present, it's not a big problem.

(NB: I am not aware; if currently it is not feasible technically to only allow SOME players into an arena - without having to invite them, in order to reduce the workload! -, Space will have to develop that feature within one year )

4 other players (Nr. 9 and Nr. 10 in each arena A and B) will be put on a reserve list (to address the problem of no-shows in the Stage 2). They will have to be present during the semi-final if they want to have a chance to play.

STAGE 3 : Semi-final

16 players are put into 4 groups of 4, classically, and play until someone reaches THREE victories, and thus qualifies for the final. Not less.

In case of total or partial no-show, the player is replaced by a reserve player; if this reserve player manages to win 3 times himself, of course he passes to the final. NB: if for instance 3 games have been played, a player disconnects (power shortage for instance) and is replaced, but comes back for the next game, he is allowed to play again.

After the stage 3, 4 other players (Nr. 2 of each group!) will be put on a reserve list (to address the problem of no-shows in the Stage 1). They will have to be present during the final if they want to have a chance to play.

STAGE 4 : Final

4 players play together until someone reaches FIVE victories, and thus qualifies for the final. Not less.

Similarly to the previous stage, in case of total or partial no-show, the player is replaced by a reserve player; if this reserve player manages to win 5 times himself, of course he is declared champion. NB: if for instance 3 games have been played, a player disconnects (power shortage for instance) and is replaced, but comes back for the next game, he is allowed to play again.

I firmly believe that with such a system:

- there are no more distorsions,

- luck factor is definitely eliminated,

- everybody is given a chance,

- we are sure to have the best 16 players in the semi-final and the best 4 players in the final,

- there is not too much work for the organisers as most of the selection is still done during the arenas, but without the current defects.

I like the idea of doing on weekends. Lots of us may be busy in the weekdays, but one problem is that we are also very busy on weekends.

I think @Indipendenza 's suggestion is sound, could do with some tweaks to simplify it a bit but the idea and approach is roughly inline what I personally think a good methodology would be. I'm sure Luke would be happy to discuss this further after the current championship.

There are many things I like about this. The first time I saw this, I was stupid enough to skim through to the conclusion. This is what I like:

- Top ten get a bye to stage two.

- Pretty much everyone has a chance.

- The arenas seem to favor nearly everyone time zone. (Right now they heavily favor Americans).

- 3 wins for the semis and 5 for the final

Things I would change:

- 4 hours seems unnecessarily to long, I don't think people want to fry their brains. I think 3 is enough because if lose a 40+ minute game you still should have enough time to come back. And also the top seven receive points so a shorter arena shouldn't be a problem.

- The top three scores of a player should be added, but a player does not have to play three arenas if he thinks he can qualify with two scores. 

- There should be four arenas for stage two, and the top four of each move on to the semis. Maybe make the arenas three or four hours. 

You have some good ideas, and its great that you are trying to eliminate the luck factors, but I think the only thing that matters is the number of games it takes to win a semifinal or a final, since good players are qualifying from the current system, and there are no noobs.

Yes, the main purpose is

a) to eliminate the luck factor 

b) to avoid a situation when an excellent player can't qualify, being 2nd X times (happened in 2021 and maybe will happen again this year),

c) to allow the best ones to be directly put further in order to make it more enjoyable and less frustrating for the players who are still less brilliant.

they should have rating groups

so people will sandbag...

This man has the perfect vision for the next tournament, I hope they actually do this. The idea of it being open for all excites me.

This has a FIFA World Cup feel to it, I like it!

It's how a World Championship should be organized.

Also, icystun is only 2300 in FFA as far as I am aware, so probably not in the top 10, but definitely 3000+ level, no?

We will see.

Also, with the current format what I found personally unfair, that's the fact the ORDER of your victories influence the outcome. That shouldn't be the case clearly, and it added a huge luck factor to something that ALREADY was influenced by luck a lot. I know at least one player who could even qualify thanks to that.

In addition, to punish by a big -2 a player who accidentally aborted the game is not fair. I know one player who, being deaf, didn't hear the signal and thus had -2, and his miraculous qualification could've been destroyed by these -2 if another player (EyeOfTiger) won just one game more (which is a piece of cake for that excellent player).

I read a comment by @LiquidFyre, saying that more people would resign games in bad positions if were not for winning streaks. (Paraphrased). I think that is a good reason for streaks in the arenas.

I do not think so.

The issue with the arenas, among other, is the TIME factor. Basically sometimes you have not enough games simply to win, even if you are very good. I could see several cases where very strong players couldn't qualify because they were stuck in long games. That shouldn't be the case. I mean, ANY format should help to designate simply the best player. Not the luckiest and not the one who could, thanks to circumstances, play enough games rather than being blocked. In one of the arenas I could only play... THREE games (and believe me, was active during all 120 minutes). Is it normal?!!! Even if I won all three of them (was not the case lol, far from that), I wouldn't have qualified. So there is a clear problem here.

Could there be 3 rating classes ? 2200 and up. Second 1901 - 2199 . Third 1601 - 1900. All 3 are run the same way. Could you possibly eliminate sandbagging by reviewing each player's rating for the last ( for example ) 6 months ? If a sudden drop is detected, it's suspicious.  

I don’t think that would make sense, especially that it naturally doesn’t involve players lower than master level. However, sandbagging removal is a good idea.