475
edits
Tournament blocks: Difference between revisions
Jump to navigation
Jump to search
Rework, partial overhaul, focusing more on the tournament aspect and less on the league.
m (more wip) |
(Rework, partial overhaul, focusing more on the tournament aspect and less on the league.) |
||
| Line 1: | Line 1: | ||
This article is meant as a repository for tournament and competitive blocks used for seating arragements. The content is mainly present in tabular form: reading it for the sake of reading is not recommended. | This article is meant as a repository for tournament and competitive blocks used for seating arragements. The content is mainly present in tabular form: reading it for the sake of reading is not recommended. | ||
== Ideal blocks == | == Ideal tournament blocks == | ||
Ideal blocks require the organizer to plan according to the situation hey are planning for | Ideal blocks require the organizer to plan according to the situation hey are planning for. | ||
=== Tournament === | === Tournament criteria to consider === | ||
Tournaments often require a single hanchan between 4 players and then a shuffling of new opponents, aiming to complete a maximum or a set number of hanchan with minimal or no repeats between players. | Tournaments often require a single hanchan between 4 players and then a shuffling of new opponents, aiming to complete a maximum or a set number of hanchan with minimal or no repeats between players. That being said, there are imperatives making it impossible to satisfy every need at once. | ||
=== | ==== Opponent mixity ==== | ||
The '''zero principle''': Never assume that your unique circumstances are the template for a fair tournament. Always assume that the model should work in all circumstances, not just tournaments of a specific length. | |||
The '''first principle''': Maximize the player base that any individual can play against. This implies one of the following: | |||
* With many players (P >= 6*R): The goal is to have people play against everyone once, with the rest at zero. No repeats. | |||
* With few players (P < 6*R): The goal is to try to have every player play with everyone else once, with a minimal amount of repeats. Gaps may occur, with some players never playing against each other, a best fit is sufficient, as it is not always possible to make an optimal setup. | |||
The '''second principle''': A tournament should present an equitable but not necessarily equal state at every stage in the tournament, with future stages building on preceding ones, contributing to a balanced state. | |||
* If (R % 4) != 0; then ''maxWind - minWind = 1''. | |||
* If (R % 4) = 0; then ''maxWind - minWind'' can take two values: either ''zero'', giving a wind balance after 4 matches (e.g.: 1111, 2222...), <br />or ''two'', with an imbalance. Note that the other two wind counts MUST BE EQUAL to R/4 (e.g.: 2110 or 0121, 1322 or 3221...). Wind counts such as {2200} are unacceptable. | |||
* If (R % 4) = 1 AND R >= 5; then any wind imbalance must be corrected at this round. {2110} must become {2111} for any concerned player. Players who had a balanced wind count in a round that was a multiple of 4 can be anything, they do not have to repeat the previous order. | |||
* A player can play the same wind back-to-back as long as the above rules are enforced. | |||
The '''third principle''': A tournament grid should be acceptable regardless of the length of the tournament. Checks and balances should occur at every round when possible before proceeding with the next. A player who is caught with a number of front-loaded matches as the same wind or north-heavy should be assured that the tournament will give him a fair chance throughout the competition. Otherwise, the tournament organizer is technically responsible for inducing players in tilt or runaway positions. | |||
The '''fourth principle''': Players should not play on the same table an abnormal amount of times. Empirically, it shouldn't exceed one of every 3 rounds. | |||
The '''final principle''': Never assume that human intervention will "self-correct" one of the system's flaws. It won't. An example would be the fourth principle above: For multi-day tournaments, the tiles could be picked up, taken away, and replaced the next day. This not only does not guarantee that the same sets can't go back to the same table, if the loading and unloading are done in a consistent manner (by human reflex), there is a much greater chance that the same sets will return to the same location. | |||
== Pitfalls and criticism of tournament blocks used for mahjong == | == Pitfalls and criticism of tournament blocks used for mahjong == | ||
=== Unfair draws leading to player tier protection or segregation === | |||
Prior to drawing lots, everyone should have the same chance of facing off against any other opponent. Lots must be established by one of two ways: | |||
* A computerized draw witnessed by many assigning all players with no preference, or | |||
* Lots must be drawn at a specific time that anyone can reasonably assist. | |||
The other mathematically unobservable fact resides in the truth that mahjong tiles can develop scratches and other identifying marks. Simply rotating players by [x,x,x,x]:->[x,x+1,x+2,x+3] will leave a quarter of the table with a potential opportunity to learn and exploit the tile markings. Luckily, the easiest way to minimize that impact is to change the transformation posited above to [x,x,x,x]:->[x+1,x+2,x+3,x+4]. | Lots cannot be half-drawn prior to the main event at a side event without players' knowledge, nor should it be done at an "optional event" that caters to only a specific in-group of people. | ||
=== Guest player exception === | |||
It is permissible to separate players from a guest organization (e.g.: Japanese pros). However, this exception cannot be used simply to split players by country, club or by other nefarious methods of discrimination. This is mahjong, not the UN. The only time it could be viewed as acceptable is during a specific inter-club rally, or at an event where each faction has equal representation, none of which has occured in NA/EU riichi mahjong. It is never acceptable to do it to avoid one specific player (the top gun, their spouse, their parent or child) and tournament organizers should show enough integrity to not allow or perform such tricks. | |||
=== Same-table issues with marked tiles === | |||
The other mathematically unobservable fact resides in the truth that mahjong tiles can develop scratches and other identifying marks. Simply rotating players by [x,x,x,x]:->[x,x+1,x+2,x+3] will leave a quarter of the table with a potential opportunity to learn and exploit the tile markings. Luckily, the easiest way to minimize that impact is to change the transformation posited above to [x,x,x,x]:->[x+1,x+2,x+3,x+4] for Dutch cycle blocks. Other tournament blocks must show some care for this. | |||
EMA tournaments have tried implementing in some of their events a "final round" where players are then re-ranked 1,2,3,4; 5,6,7,8; et cetera, enabling them to have 8 hanchan in an event where 24 or 48 people show up, as it is mathematically difficult or impossible to satisfy for more than 7. Otherwise, most of their tournament seating arrangements occur though Dutch cycles (transparent but with risks), or through software which may or may not be easily verifiable or reproducible (trusting the black box, security through obscurity). However, they have not statuated on a method for running tournaments. The two main software solutions have benefits but have significant drawbacks. Criticism for these kinds of points in participative tournaments is not generally a thing in Japan, as most tournaments there are knock-out events or small groups that accept duplication of pairings. | EMA tournaments have tried implementing in some of their events a "final round" where players are then re-ranked 1,2,3,4; 5,6,7,8; et cetera, enabling them to have 8 hanchan in an event where 24 or 48 people show up, as it is mathematically difficult or impossible to satisfy for more than 7. Otherwise, most of their tournament seating arrangements occur though Dutch cycles (transparent but with risks), or through software which may or may not be easily verifiable or reproducible (trusting the black box, security through obscurity). However, they have not statuated on a method for running tournaments. The two main software solutions have benefits but have significant drawbacks. Criticism for these kinds of points in participative tournaments is not generally a thing in Japan, as most tournaments there are knock-out events or small groups that accept duplication of pairings. | ||
| Line 30: | Line 55: | ||
| 1 || 1 || 01 || 12 || 23 || 34 | | 1 || 1 || 01 || 12 || 23 || 34 | ||
|- | |- | ||
| 2 || 2 || 13 || 01 || | | 2 || 2 || 13 || 01 || 37 || 25 | ||
|- | |- | ||
| 3 || 3 || 27 || 40 || 01 || 14 | | 3 || 3 || 27 || 40 || 01 || 14 | ||
| Line 37: | Line 62: | ||
|- | |- | ||
| 5 || 5 || 01 || 16 || 31 || 35 | | 5 || 5 || 01 || 16 || 31 || 35 | ||
|- | |||
| 6 || 6 || 17 || 01 || 38 || 33 | |||
|- | |||
| 7 || 7 || 24 || 41 || 01 || 18 | |||
|- | |||
| 8 || 8 || 44 || 26 || 19 || 01 | |||
|- | |||
| 9 || 9 || 01 || 20 || 28 || 36 | |||
|- | |||
| 10 || 10 || 21 || 01 || 39 || 30 | |||
|- | |||
| 11 || 11 || 32 || 42 || 01 || 22 | |||
|- | |- | ||
|colspan=6| '''Notes:''' This is a sample of the table matching for a 44-person tournament. As players are evenly divided in 4 blocks of shifting players, when a block exceeds the size allotted, players shuffle back. E.g.: On table 5 above, the 4th block comprises of playres 34 to 44. One group is moving forward in steps of 3, and 43+3 is 46. Naturally, there is no player 46 to take that seat and is filled with the 2nd player (44+2) of that block, player 35.<br/> | |colspan=6| '''Notes:''' This is a sample of the table matching for a 44-person tournament. As players are evenly divided in 4 blocks of shifting players, when a block exceeds the size allotted, players shuffle back. E.g.: On table 5 above, the 4th block comprises of playres 34 to 44. One group is moving forward in steps of 3, and 43+3 is 46. Naturally, there is no player 46 to take that seat and is filled with the 2nd player (44+2) of that block, player 35.<br/> | ||
| Line 253: | Line 290: | ||
=== League play blocks === | === League play blocks === | ||
This section may receive heavy eits or be deleted in the future. It is not worth the effort to consider the league issues until a far off future. | |||
Considering that league blocks contain 5 players, the counting mechanism has to be recalculated almost from scratch. These numbers are to satisfy 6-session events or seasons. For all Dutch cycles, as well as some non-Dutch SGP blocks can drop the last fifth in order to make groups of 4, expanding the solved ranges of players from as low as 80% to 100% of the maximal solutions. Considering all our solutions for 25+ are good for 6+ sessions, solutions for larger numbers can concatenate smaller groups with the minimum amount of sessions needed to make a larger group that satisfies that lower bound. The solution for 25 players in 6 sessions can drop one player, the solution for 40 present below can drop 5, although it may be possible that a solution allowing to drop 8 exists. | Considering that league blocks contain 5 players, the counting mechanism has to be recalculated almost from scratch. These numbers are to satisfy 6-session events or seasons. For all Dutch cycles, as well as some non-Dutch SGP blocks can drop the last fifth in order to make groups of 4, expanding the solved ranges of players from as low as 80% to 100% of the maximal solutions. Considering all our solutions for 25+ are good for 6+ sessions, solutions for larger numbers can concatenate smaller groups with the minimum amount of sessions needed to make a larger group that satisfies that lower bound. The solution for 25 players in 6 sessions can drop one player, the solution for 40 present below can drop 5, although it may be possible that a solution allowing to drop 8 exists. | ||