I propose the rotating blocks system. Read about it here:
mukilteochess.pbworks.com/w/page … 0explained
This is an attempt to improve the Mini-Swiss at Quad tournaments. The main problem with the mini-swiss is that players who receive full-point byes go on to win the mini swiss, when other players actually performed much better.
This tournament format was discussed in the topic “40/25, SD/10+5” starting at post #299831.
Yes, and I created a new thread because it would be off topic to discuss the rotating blocks in a section discussing a specific time control.
Welcome, Travis, to the world of live posting! I’m glad you finally jumped in here to represent yourself directly.
I note that your 10-4-2015 event did not have enough players, nor the right division of players, for you to try out your rotating triangle method.
All others who have been following this (and the other) thread, please note that “rotating triangles” is simply a new way of sectioning a quad event when the total number of players is a multiple of 2 but not a multiple of 4.
The usual method, the 6-player Swiss, creates a single enlarged section for the bottom 6 players. The rotating triangle method, by contrast, creates two sections of 3 players each. The 3 players in each section all play each other, and each plays a third game against one of the players in the other section.
The idea is simple enough that a fancy name like “rotating triangles” may serve only to obscure the basic idea, and may have caused some posters here to lash out at a new idea that, in my opinion, merits serious consideration.
One advantage of the two 3-player sections is that, in a quad event with prizes, there can be prizes for each of the two 3-player sections. These prizes can, if desired, be based solely on the two games each player played within his own section, with the cross-section games counting for ratings but not for prizes.
For the above reason, I do have a quibble with the colors you suggested in your original paper. It seems to me it would highly desirable for each player to be assigned 1 white and 1 black in the 2 games played within his own section. Your proposed color scheme does not accomplish this; mine does. (See that other thread.)
Incidentally, I don’t quite understand why you mentioned full-point byes in your initial post in this thread. Apparently your solution to byes is to jump in as a player into the event you are directing, which is a laudable solution. Then you will have an even number of players. The rotating triangles – oops, the two 3-player sections – deals only with the case where the number of players is even, but not a multiple of 4, and that’s all it needs to do.
Bill Smythe
Welcome to the forum, and thank you for running events.
That said, your method has a serious deficiency.
The Swiss becomes two round robins with a round of interleague play. A Swiss System is preferred because it reasonably approximates the result of a round Robin tournament in a limited number of rounds. Your method does nothing of the sort, and in fact contaminates the results of one of the round robins with an extra game outside the round robin.
The only benefit is a fixed schedule determined early.
I disagree.
What could be better than having two 3-player sections rather than one 6-player section? Each 3-player section then becomes a full round robin – closer to the experience enjoyed by the players in the upper sections.
As for the word “contaminates”, it is ludicrous to use such a strong word to describe the effect of extra games. Extra-game sections are used all the time in Swiss events, to accommodate players who would otherwise receive a full-point bye or a forfeit win. It is a simple courtesy to the players. So why not do it here too?
If the quad event has prizes, the organizer can simply declare that only the intra-quad games will count for prizes. That would totally eliminate any trace of “contamination” that any reasonable person could possibly claim exists.
I don’t know what kind of wall charts are posted at Travis’s events. If they are in the form of standard round-robin charts (with black boxes going down the main diagonal and all that good stuff), then the 3-player sections can be posted that way too, as 2 separate sections, along with a separate (Swiss-style) chart for the “extra games”, i.e. for the inter-quad games.
And please don’t minimize the benefit of “a fixed schedule determined early”. That benefit is yet another way to make the players feel more welcome, as though they are playing in the same tournament as the players in the upper sections.
Bill Smythe
This strikes me as being a sort of reversed Swiss: In a Swiss, you begin by pairing upper half against lower half, usually with expected outcomes peppered by occasional surprises, and then gradually refine the ordering of winners and losers by pairing winners of prior rounds against winners, and losers of prior rounds against losers. But in rotating blocks, you don’t pair upper half against lower half until the end.
The fairest way to run a tournament is to have a single Round Robin in which all players participate. But this is usually impractical because it would require a ridiculous number of rounds. The Swiss is an attempt to choose an absolute winner as fairly as can be done with a reasonable number of rounds, but what usually seems to happen in practice is that Swiss tournaments are divided into multiple sections (usually based on rating, though in scholastic tournaments, they may be divided based on grade or age). Quads, on the other hand, abandon any pretense of being able to choose an absolute winner, and begin by automatically dividing players into rating-based sections of a small, predetermined size. So the question is whether the intent of Quads is better satisfied, when the number of players is not evenly divisible by 4, by having a single larger-sized group on the bottom, or by having two smaller-size groups. My preference is for the first of these.
Bob
The intent of Quads is to minimize rating differences among the players who will be playing each other. Therefore, two 3-player sections is preferable to one 6-player section, because then at least 2 of each player’s 3 opponents will likely be close to that player in rating.
Bill Smythe
I don’t understand. Don’t they do exactly what you claimed was ideal, that is put all players into a single round robin?
Alex Relyea
Nope. If you have 20 players, there is a big difference between a 20-player (19-round) round robin and five 4-player (3-round) round robins.
Bill Smythe
Like interlocking gears.