A Look at Different Tournament Systems

slowpoke-firered-leafgreen-rg-80

Scholarly Slowpoke.

Before I start, I would like to thank Adam for letting me write an article for the SixPrizes front page. I have recently started a blog, Pokedemia, where I publish articles about Pokémon with a more academic approach.

Besides publishing articles, tools to help Pokémon TCG players will also be released as they get developed. Players also have the opportunity to write articles for the blog if they wish. You can find a longer presentation of myself on there, so I would like to keep it short here.

My name is Steffen Eriksen and I’m from Denmark, but currently living in the Netherlands. I have been playing the Pokémon TCG since 2005, with my best result being top 16 with Flygon/Nidoqueen at Worlds 2007. Besides playing competitively, I have also spent a lot of my time judging and organizing events.

But enough about me; let’s get to the topic at hand: alternative tournament systems.

What are Tournament Systems?

All Pokémon TCG players nowadays are used to playing a tournament system where they play a certain number of Swiss rounds followed by a top cut, which then is run using Single Elimination. This article will discuss the pros and cons of alternative tournament styles which could be used instead of the Swiss plus Single Elimination system which the Tournament Operations Manager (TOM) supports.

The question of which tournament styles to use has been frequently debated and most recently on The PokeGym. Some argue that most games in the current system are pointless and without meaning and that after 1, 2, or 3 losses (depending on the size of the tournament) it is not worth it to keep playing, as it will be impossible to reach top cut. Others will play it out because it is fun or for the spirit of the game.

In addition, dropping out from a tournament will affect your opponent’s resistance negatively and thereby make it harder for them to reach top cut if it comes down to tiebreakers. (For a discussion about the chances of reaching top cut at tournaments see this article.)

The problem of missing out top cut due to tiebreaker is a frequently seen problem with the current system as you cannot control the outcome of your opponent’s games. Possible solutions to this issue will be discussed later in this article. Designs which will favor the best ranked teams going into the finals will also be reviewed.

There exist a lot of different tournament styles and since it would be too time consuming to treat all of them, only a small selection of tournament styles will be treated in this article.

The Current System

top cut bracket standings

A tournament system from the anime.

Before discussing alternative tournament styles it would be good to give a short description of our current tournament style. TPCi just announced changes for the new season and some of them will affect the current Swiss plus Single Elimination system as we know it. For larger tournaments (States and above) there will be 50 minute best-of-three Swiss rounds and a maximum top 8 cut. Although this has no influence on the way the Swiss system works in general, it will have influence on the number of Swiss rounds played and the size of the top cut.

The currently approved tournament styles include: Swiss, Single Elimination, or Swiss plus Single Elimination. Throughout this article a hypothetical tournament using Swiss plus Single Elimination, containing 32 Masters players, will be considered. (This is done just to keep the discussion as general as possible. In addition it will be assumed that each round of Swiss lasts 30 minutes and each Single Elimination round lasts 60 minutes).

The size of the top cut of this hypothetical event would not be affected by the change in the top cut structure as announced by TPCi. For the first round of the tournament players are paired randomly against each other and receive 3 match points for winning a game, 1 for an intentional draw, and 0 for losing. In the following rounds, players with the same amount of match points are paired randomly.

In the case of an odd number of players having a certain record, one of the players is randomly paired down. If the group with the lowest match points (0-X) contains an odd number of players, one of the players is randomly assigned a bye. This process continues until only 1 player remains undefeated. (With drop outs it is possible that there will be no undefeated player, i.e. the undefeated player drops out.)

After running an appropriate number of Swiss rounds, according to the number of players, the top ranked players are seeded into Single Elimination brackets, based on their Swiss ranking, and play until one player remains, who is then the winner of the tournament.

Guidelines for determining the appropriate number of Swiss rounds and size of the top cut, can be found in the appendix of the original article.

In order to determine the final placement after the end of the final Swiss round, a number of tiebreakers as described by the rules and resources should be applied. These tiebreakers include: number of match points, opponents’ win percentage, opponents’ opponents’ win percentage, head to head, and finally the standing of the last opponent. For more detailed information about tiebreakers, tournament procedures when using Swiss plus Single Elimination, click here.

More Flights?

flying-pikachu-wizards-black-star-promos-25

Not that kind of flight.

Before discussing completely different tournament styles, there is a small extension to the current system which could be considered. In the tournament procedures, as described here, it is allowable to create different flights if the number of players exceed 64. This option may currently only be used as a last resort if it would otherwise be impossible to finish the tournament before the venue closes.

(Flights have to be used if the number of players in one age division exceeds 512.)

Increasing the number of flights reduces the number of Swiss rounds. The number of flights must be 2, 4, 8, 16… and for each increase in the number of flights, the number of Swiss rounds is reduced by 1.

It is not allowable to have more than 1 flight in the example tournament considered here with 32 players, but imagine now that it was possible. Having 2 flights would then create two pods with 16 players each, and each pod having a top 4, in contrast to 1 flight with 32 players and top 8. The number of Swiss rounds would then be reduced from 5 to 4, saving 30 minutes of effective playing time.

If this process were to be repeated, so there would then be 4 flights, each pod would then consist of 8 people. The number of Swiss rounds would then be further reduced to 3 and each pod would have a top 2. Continuing this process to the extreme case, with 16 flights, would result in a Single Elimination tournament.

Let us consider the case where there are 8 flights. Each flight now contains 4 people and the number of Swiss round would then be 2, following the process from before. Comparing to the original 1 flight tournament, the effective playing time is reduced by 90 minutes. However, with such a low pod size, a Round Robin could be considered, increasing the number of preliminary rounds from 2 to 3.

This slight variation of the current system would save 60 minutes in effective playing time and allow everyone who is in the same flight to play against each other. This variation would, however, not solve the problem about missing top cut due to tiebreaker. In addition, having so small flights also reduces the number of losses a player can afford before being eliminated from the tournament.

Losing 1 game in a flight with only 4 players and top 1 can eliminate you from the tournament, whereas 1 loss in a flight with 32 players and top 8 would not cost you the tournament, as you would still be secured a spot in the top cut.

A reason for not allowing flights sizes below 32 players could be the fact that then just 1 loss could eliminate a player from top cut. (Indeed it is possible that in a tournament with only top 4, 1 loss can eliminate a player from the tournament due to tiebreaker.)

Alternative Tournament Systems

After viewing a simple extension of the current tournament system, it is now time to take a look at completely different tournament systems. Some of the systems discussed below might be similar in nature to the current system, but each one of them has a certain distinction.

Round Robin was briefly touched upon in the previous section, but this structure would only be possible to implement when the size of the tournament is very small. Round Robin would otherwise be ideal to identify the winner of the tournament, since each player has to face all the other players in the tournament and the overall best player would then have been identified. With 32 players, 31 rounds would be needed to determine the winner. This tournament would be unrealistic to finish within a reasonable time frame.

Double Elimination

here-comes-team-rocket-team-rocket-tr-15

Prepare for trouble… and make it double!

The first completely different tournament system which will be discussed is Double Elimination (DE). Unlike Single Elimination (SE), a player can still lose one game and win the tournament.

A DE makes the use of two brackets; the winners bracket and the losers bracket. All players start in the winners bracket and when a player loses, that player drops down from the winners bracket to the losers bracket. If a player playing in the losers bracket is defeated, that player is removed from the tournament.

DE is currently used in various computer gaming tournaments, table football and the NCAA baseball tournament.

The tournament can either be conducted as a “blind draw,” where each player randomly receives his or her seat in the bracket, or seeded where each player is ranked into the elimination bracket based on some predetermined measure. (In the case of Pokémon TCG, this could for instance be a player’s Elo ranking or Championship Points.)

The maximum number of games (not rounds) played in a DE tournament is one less than twice the number of players attending and the minimum number of games is two less than twice the number of players attending. In the example used here the maximum and minimum number of games would be 63 (32 x 2 – 1) and 62 (32 x 2 – 2) games.

The reason why the number of games can vary is that in the finals the winner of the winners bracket faces off against the winner of the losers bracket, and if the winner of the losers bracket would win the first game it would be unfair to eliminate the winner of the winners bracket after his or her first loss. Therefore the winner of the losers bracket has to win twice over the winner of the winners bracket in order to claim the title as the winner of the event.

An example of a DE bracket (with seeding) can be seen in the appendix of the original article.

One advantage of DE over SE is that the third and fourth place finishers can be determined without having a classification match between the two. Seeding in DE (and SE) is used to prevent the strongest players from meeting each other in the early rounds. If a DE tournament would be run as a blind draw it still can overcome a problem seen in SE, where a player is eliminated after his or her first loss, since the loser can still fight his or her way through the losers bracket to the finals and win the event.

A disadvantage of DE over SE is that at least twice the amount of games has to be conducted, as players have to be defeated twice to be eliminated from the event (with N players in the event, a total of 2N-1 or 2N-2 games have to be conducted, as discussed earlier).

Depending on the point of view, DE can have an advantage (or disadvantage) over the standard Swiss format employed in the Pokemon TCG, in that it removes many meaningless games by effectively removing players from the tournament who have lost their chance of winning. In addition it could then be argued that it would raise the competitive level at tournaments.

Many players play not only to winning, but also for the fun. As the Pokemon TCG’s main target group is children, it might not be wise to employ a highly competitive tournament system, as this would discourage many kids from playing. Using the Swiss system, it does not enforce such a competitive environment as DE, since all players can play a minimum of 3-9 games (depending on the size of the tournament), which in any case is more games than guaranteed by DE.

However, there is an easy extension to DE which ensures a minimum of 3 games for all players.

A Conclusion… For Now!

This concludes my coverage of the Double Elimination tournament system. If you wish to read about other tournament systems and the possibility of implementing a seeding system, then view the full article (and others at Pokedemia). Even the full article does not cover all the different possibilities that exist. It is a broad subject of which there is still much to be written.

I hope that you liked my article and feel free to comment below. Was it too long? Is there any specific tournament system you feel should be discussed? Again, thanks to Adam for giving me the opportunity to write this article and good luck to all of you in the upcoming season!

Steffen Eriksen

  • Obro

    You should also mention that you placed 33rd at ECC this year ;p

    • Sauce_Boss

      Surely no one would like reminding of coming 33rd at the ECC…? Eh, @hunkyross?

      • Ross Gilbert

        Hey, I’m completely over that…. I mean, 6-2 in a 240 person tournament shouldn’t guarantee cut right…. :’(

        • SuperE_DK

          Oh well… It really hurts your tiebreaker when you lose the first game and your opponent later drops… Then going 6-2 is often not enough. I think that it should change a little, so that your tiebreaker is not destroyed by your opponents dropping out.
          I’m also over it, but at the moment it was very annoying!

  • SuperE_DK

    Btw. Uploaded my follow up article, where I look at a possible way the new large tournament could be conducted, with regards to the 2nd day Swiss rounds. Check it out:
    http://www.pokedemia.com/2013/08/the-new-tournament-system.html

  • Scenia

    Why is there nothing about the system used commonly in Japan? I think that would have been interesting for many readers.
    For instance, this system consists of any number of rounds of “Open Playing” where players have to win several times in a row to earn a spot in the top X. Once they lose, they can re-register and start over. Once all spots are taken, SE Top Cut starts (and players can switch decks for Top Cut). I have no idea what it’s called, though…

    Don’t get me wrong, I think this system is flawed in many ways (one being that it totally destroys slow decks so that someone can go undefeated but still not earn a spot), but it would have been nice to see it discussed either here or in the original article.