Results 1 to 7 of 7

Thread: Article: Tournament Formats for HEMA / WMA

  1. #1
    Join Date
    Apr 2003
    Location
    Belgium
    Posts
    732

    Article: Tournament Formats for HEMA / WMA

    Tournament Formats for HEMA / WMA
    By Matt Galas, Copyright 2010

    This is another in a series of posts about tournament issues affecting the HEMA / WMA community. This post will focus on tournament formats. While it's fun to argue about particular rules (double hits, after-blows, grappling, etc.), the question of tournament format is far more important in terms of the fighters' experience of the event. If the tournament organizers don't have a clear idea of how to organize and format a tournament, the end result will be chaos, and nothing else will matter.

    To be clear, the term "tournament format" as used in this post refers to the structure of the tournament. In particular, it refers to how the fighters are paired up, how many fights they get, how they progress towards the end result, and how the eventual winner is determined. The most common types of tournament are single elimination; double elimination; pool tournaments (round robin or otherwise); swiss pairing; and combinations of pool and single/double elimination. Another type of format, based on the old Franco-Belgian tournament rules, is the "King of the Hill" format. Each of these will be discussed in detail further below.

    Before discussing specific formats, it is important to understand that there are several competing goals that must be considered, since the choice of format amounts to a series of trade-offs between these competing concerns. These concerns are the following:

    - Time management (mainly affected by the number of fencers and the amount of time available)
    - Ease of administration (linked to time, plus the numbers, experience, and skill of judging staff)
    - Fighter satisfaction (linked in some degree to number of guaranteed fights)
    - Audience satisfaction (linked to existence of a climactic "final")
    - Achievement of community goals (such as an objective evaluation of a fighter's skill)

    The First Question: To Seed or Not To Seed

    In any tournament, the first question to be resolved is how to match the fighters. Should they be paired at random, or according to some criteria? Random assignment of fights may seem to be fairest at first glance, but this results in certain problems. For example, random pairing of fighters can result in the best fighters being matched up early in the tournament. This means that the best fighters can be eliminated early, leaving less-skilled fighters to battle for the final. This increases the role of chance in the end result, and lessens the likelihood that the tournament winner is actually the best fighter. This leads to a tournament that may well be unsatisfactory to the audience, the tournament organizers, and many of the fighters.

    One way to resolve this problem is by resorting to different forms of seeding; that is, by matching up fighters according to various criteria. As an experienced martial arts coach once remarked to me, "If you ever get to the finals of a tournament, and wonder how the hell these two clowns ended up at the top -- it's probably because you didn't use seeding." Here are the most common ways of matching up fighters:

    Best-Worst Seeding:

    Most commonly, seeding takes the form of pitting the best ranked fencers against the worst. Assuming 10 fighters, number 1 fights number 10; number 2 fights number 9, number 3 fights number 8, and so on. Statistically, this yields a high likelihood of the best fighters reaching the finals. It creates a better tournament from the perspective of the audience, since the early fights (mismatches between "very good" and "very bad") give great scope for the better fencers to show off their skills, while the later fights are exciting matches between evenly-matched fencers of greater skill.

    On the other hand, this kind of seeding can be viewed as unfair to new fighters, who face an uphill battle that makes it very unlikely that they could ever win the tournament. It can also be seen as unfairly favoring the most experienced fighters, who face a series of easy opponents, making their road to the finals much smoother. Despite these concerns, this type of seeding is commonly used in competitive sports.

    Like-Like Seeding:


    This method of seeding, used in the Swiss Pairing format (see below), matches fencers with opponents of equal skill. As the tournament progresses, this form of seeding is used to ensure that each fencer is matched with a challenging competitor with a similar win-loss ratio. As opposed to Best-Worst seeding, this is fair to all fencers, since it provides a meaningful but challenging experience for all levels of competitor. It also results in a true ranking of competitors in terms of their performance on that day. While this ranking is accurate at the upper and lower ends of the scale, it is less accurate and reliable for those in the middle. It is also very complicated to administer, although software designed for this purpose can make this easier. This type of seeding is used in international chess competitions.

    Top Half vs. Bottom Half:

    This method of seeding seeks to split the difference between the forms listed above. Assuming 10 fencers, the top half (numbers 1-5) are matched against the bottom half (fencers 6-10). Thus, number 1 is matched with number 6; number 2 is matched with number 7; number 3 is matched with number 8; and so on. This method has the effect of matching the top 50% against the bottom 50% in a manner that the skill differential between fencers in each pair is roughly the same. This has the benefits of Best-Worst seeding, but avoids glaring overmatches, and limits the degree of unfairness to junior competitors. This method of seeding is used in international fencing competitions.

    Other Seeding Criteria:

    Other criteria than fencer quality can be applied to tournament pairing. For example, the recent tournament at Apelern, Germany used seeding based on familiarity among the fighters. Fencers were placed into pools according to club affiliation and similar factors, to ensure that each fighter fought his bouts against unfamiliar opponents. This increased the quality of the tournament for the fighters, who were guaranteed a certain number of fights against new adversaries.

    With the considerations above as a backdrop, we now move on to a discussion of the various tournament formats.
    Last edited by Matt Galas; 12-07-2010 at 01:49 PM.
    Matt Galas
    Mons, Belgium

  2. #2
    Join Date
    Apr 2003
    Location
    Belgium
    Posts
    732
    Single Elimination Format:

    In this format, each fencer is matched with an opponent (either randomly or by seeding). The loser is eliminated from the tournament. The winner proceeds to the next round, where the process repeats itself. This continues until the final, where the two previously-undefeated fencers confront each other. This results in a "pyramid" diagram which makes the progression very clear. (Click this link to a diagram and a more detailed explanation: http://en.wikipedia.org/wiki/Single-...ion_tournament )

    Pros: This format is fast. It allows a large number of fencers to fight, yet rapidly selects a small group of winners to proceed towards the final. It is also very easy to administer. From an audience perspective, it is easiest to understand, since the losers are simply eliminated. For these reasons, this format is ideal for an event where the tournament is merely an added attraction, and not the main event.

    Cons: This format is the worst for the fighters, since they are only guaranteed a single fight. It also increases the role of chance, since if they may be eliminated because they are matched with a much superior opponent, get a bad call from the judges, are slow to warm up, or just happen to have a bad first fight. Considering the scarcity of well-trained judges in our community, it maximizes the role of poor judging, which other formats compensate for. The unforgiving nature of this format (screw up once, and you're done) is also more likely to create a feeling among fighters and the audience that the result was unfair. This format places maximum emphasis on winning at any cost, since any failure means elimination. Finally, it does a poor job of ranking the fighters in terms of relative skill levels, which other formats (especially pool fencing) do much better.

    Double Elimination Format:

    In this format, each fencer is matched with an opponent (either randomly or by seeding). The loser goes to the losers pool; the winner stays in the winners pool. The fencers in these two pools continue to fight. After his second defeat, a fencer is eliminated from the tournament. This results in a structure that looks like a double pyramid, with the winners of each pyramid finally meeting in the finals or semi-finals. (Click this link to a diagram and a more detailed explanation: http://en.wikipedia.org/wiki/Double-...ion_tournament )

    Pros: While not as fast as single elimination, this format is still quite fast. It allows a large number of fencers to fight, and guarantees them at least two fights. It is also fairly easy to administer. It represents the best balance between the competing concerns of time management and fighter satisfaction (at least from the perspective of the tournament organizer).

    Cons: This format still only guarantees fighters two fights. It is harder to administer, and the double-pyramid diagram is not as intuitive to understand. It still allows a fairly large role for chance and poor judging to affect the outcome. Bottom line, it has all the disadvantages of the single elimination format, only to a lesser extent.

    Round Robin:

    The easiest way to determine who is really fencing best on a particular day is to have every fencer fight every other fencer. This format, called Round Robin, is easily diagrammed and administered. (See this link for a diagram and more information: http://en.wikipedia.org/wiki/Round-robin_tournament )

    Pros: This format maximizes the number of fights, since every fighter gets the same number of fights as there are opponents. It reduces the role of chance, since every fencer is fighting the same opponents. It reduces the role of bad judging, since with such a large number of fights, the number of bad calls will be more evenly distributed among the fighters. It is also easy to administer. Finally, it is perhaps the best method for ranking fighters, since it gives the largest sample of each fencer's combative ability. Overall, it represents the best choice in terms of fighter satisfaction (at least from the perspective of number of fights).

    Cons: This format is very time consuming. It is only possible with relatively small numbers of fencers, or where there is a large amount of time at the organizer's disposal (rarely the case). It also requires enough space to run multiple fights at the same time, as well as a larger judging staff. It is unsatisfying for the audience, since it does not result in a climactic final between the "best" fighters. (This can be offset by combining this with other formats, discussed below.) It can also result in situations where all fights have not been completed, but the winner's identity is already mathematically certain (because of the number of fights won); this makes many of the later fights irrelevant, and creates the possibility that these fights will not be hotly contested by the fighters. Moreover, the Round Robin format does not really eliminate the role of chance; although every fencer fights the same opponents, they don't fight under the same conditions. (For example, I may fight a particular fencer while he is fresh, while you fight him after he has fought 10 bouts, and is tired.) It is also the most physically demanding format for the competitors, since they will not only be expending energy, but will be accumulating bruises and injuries along the way. Finally, because there is no elimination, it does not generate the kind of stress found in other formats. From a martial perspective, this is a disadvantage. Because of the large number of disadvantages listed above, this format is seldom used.

    Swiss Pairing:

    This format was originally developed for chess competitions. It is intended to yield the same ranking benefits of the Round Robin format, but in much less time. The tournament begins by matching up pairs of fencers (randomly or by seeding). After the first round of bouts, fencers are ranked according to win/loss ratio (1 or 0). Then, fencers are matched against those fencers who are closest to them in win-loss ratio and overall scoring. In the next round, the fencers are again ranked according to win-loss ratio (now, 2-0, 1-1, or 0-2). This pairing of fencers according to ability continues into subsequent rounds, until a pre-set criterion has been achieved (such as completion of 10 fights). Throughout the process, fencers only face the same opponent once.

    Pros: This format is based on a form of seeding that pairs each fencer with opponents who are at a similar skill level. Thus, it is most challenging for the fighters, while also being the fairest to them. Unlike other forms of seeding (e.g., best vs. worst), Swiss Pairing does not include a bias in favor of the best fencers, who are given a string of easy opponents, while new fencers are certain to be weeded out by a string of increasingly tough opponents. Like the Round Robin format, this format maximizes the number of fights, but in a fraction of the time. Also like the Round Robin format, it reduces the role of chance, and bad judging. It is perhaps the best method for ranking fighters, since it gives a large sample of each fencer's combative ability, paired against opponents of similar skill. Thus, Swiss Pairing would be valuable as a method for quickly generating a ranking system for tournament fighters.

    Cons: Swiss Pairing is more complicated to administer, although this can be offset by software that is readily available on the internet. It is also not immediately understandable to competitors, tournament administrators, or the audience. Like the Round Robin format, it has the disadvantage of not yielding a climactic "final", so is unsatisfying for the audience. (This can be offset by combining this with other formats, discussed below.)

    Round Robin Pools, Followed by Single or Double Elimination:


    The big disadvantage of Single and Double Elimination formats is the question of fighter dissatisfaction over the limited number of fights. No-one wants to travel to a tournament, only to be eliminated after one or two fights. One way of resolving this issue is by combining the Round Robin format with the Single or Double Elimination.

    To do this, the fencers are divided into a limited number of sub-groups, called "pools", ideally composed of the same number of fencers. (6 or 7 fencers is often considered to be best, in terms of time management and number of fights.) In each of these pools, the fencers fight "round robin", with each fencer fighting every other fencer in the pool. After the pool fencing is complete, the fencers are ranked according to various criteria (win-loss ratio, hits gained vs. hits received, etc.).

    The next phase of the tournament consists of a Single of Double Elimination format. The ranking which results from the pool fencing is used to seed fencers (best vs. worst). Alternately, the ranking can be used to determine who moves forwards into the elimination phase (for example, by advancing only the top 50% of fighters). The elimination phase (single or double) then proceeds to the finals in the usual way, described above.

    Pros: This format combines the best of the Elimination and Round Robin formats. It provides a large number of fights for the fencers, without taking the time required for a true Round Robin tournament, and without the physical toll on the competitors. It yields a climactic final, which is a major disadvantage of the Round Robin format. It represents the best balance between the competing concerns of time management and fighter satisfaction (at least from the perspective of the fighter).

    Cons: This format requires a fairly large amount of space, since it requires multiple fights to take place simultaneously. It requires a larger judging staff for the same reason. It is harder to administer, since it combines two different formats. Pools can be difficult to arrange if the number of competitors is not easily divisible.

    Swiss Pairing, Followed by Single or Double Elimination:


    Another way of resolving this fighter dissatisfaction with Single or Double Elimination is by combining this format with Swiss Pairing. To do this, follow the Swiss Pairing format described above. Once the fencers have fought a certain number of fights (or some other criterion has been achieved), the top fencers are selected to compete in a Single or Double Elimination phase. For example, the top 50% of fighters proceeds to the elimination phase, which then proceeds to the finals in the usual way, described above.

    Pros: This format combines the best of the Elimination and Swiss Pairing formats. It provides a large number of fights for the fencers. It yields a similar ranking to pure Swiss Pairing. However, it also yields a climactic final, which is a major disadvantage of the Swiss Pairing format. It represents an excellent balance between the competing concerns of time management and fighter satisfaction (at least from the perspective of the fighter).

    Cons: This format requires a fairly large amount of space, since it requires multiple fights to take place simultaneously. It requires a larger judging staff for the same reason. It is harder to administer, since it combines two different formats.

    Franco-Belgian Format ("King of the Hill"):

    This is an historical format, drawn from the prize-playing rules of the Franco-Belgian fencing guilds. It was used with longsword, rapier, and rapier & dagger from the early 16th century until the fencing guilds were dissolved in 1791 as a result of the French Revolution.

    In this format, one fencer is selected to be the "King" or "Defender". The other fencers are assigned numbers by lottery. The goal of the other fencers (the "Challengers") is to defeat the King and take his place. They do this by scoring a "clean hit" on the King. To be considered a "clean hit", the Challenger must score a hit on the valid target area (irrelevant to this discussion); must deliver the attack without a double hit; and must escape the King's "after-blow". (The King must deliver his after-blow immediately after he is hit; otherwise he loses this privilege.) Thus, the King has a major advantage over the Challenger, who must exercise consummate good swordsmanship in order to close safely, strike the King, and escape unscathed.

    Depending on the guild rules, each fencer has a limited number of tries (called "venues"), normally three, to strike the King. These either take place consecutively (with each fencer making all three tries at once), or in turns, with each fencer making one try to defeat the King, and then going to the back of the line if he fails. A fencer who succeeds in defeating the King takes his place, and enjoys the advantage of the double hit and the after-blow. This continues until all fencers have exhausted their "venues."

    Pros: Despite the unusual format, this has a number of advantages. It is easy to administer, and yields a fast-paced competition. It is very time-efficient. It sorts out the best fighters quickly, since they must have good offensive skills to hit the King, and equally good defensive skills to defend against his after-blow. It promotes good fencing, since it integrates principles of good swordsmanship into the format itself. It is historical in nature, with elements traceable not only to France and Belgium, but also to Italy and perhaps England.

    Cons: Despite its historical nature, this format is alien to modern sportive conceptions. It does not provide an even playing field for the competitors. It requires special skill sets of the judges (e.g., to judge the after-blow). It does not yield a climactic "final" like the various Elimination formats.
    Last edited by Matt Galas; 12-07-2010 at 02:00 PM.
    Matt Galas
    Mons, Belgium

  3. #3
    I think this is a great overview of tournament types. When it is feasible, I think Swiss followed by elimination is the best choice, but I find the Franco-Belgian rules surprisingly appealing as well, perhaps in part because they're historical.
    -Bradley L'Herrou

    Finding Swetnam

  4. #4
    Join Date
    Oct 2002
    Location
    Houston-Galveston, TX, USA
    Posts
    36
    Round Robin, Direct (single) Elimination and Double Elimination (repechage) are established formats used by the international sport fencing community. Forms for tracking the events are easily available.

    For Round Robin

    An example of Direct Single Elimination

    Repechage

    Additionally, because even local fencing tournaments are frequently run by computer, there is software available that recognizes these formats. One is Fencing Time. Another is XSeed

    Anyway, these are resources available for free that might be adapted for WMA/HEMA events.

  5. #5
    Join Date
    Oct 2002
    Location
    Houston-Galveston, TX, USA
    Posts
    36
    I should add that the most common system in local fencing tournaments in the US is to break up competitors in an event into pools of 6-7 fencers. These are generally seeded based upon a fencers ranking (from past events).

    For example, an event with 18 fencers would be broken into 3 pools of six. The best three fencers (based on past resuts) would be in different pools. The same for the next best three. Among fencers with no past track record classification (unranked), they are assigned randomly, with an eye to putting clubmates in different pools (Why pay money at a tournament to fence the same guys I fence every week?)

    As you can see from the paperwork, everyone in a pools fences everyone else in their pool. The results of the pools are used to seed fencers into Direct Elimination (with or without repechage).

  6. As some one I would consider as "the worst" I would not expect to win a tournament but I have always found that in just about any thing from music to fencing playing or competing against the best makes you better even if you lose you will do better the next time and you have a better idea of where your weaknesses are I just wonder how the better or "best" feel about it as I have never been on that side of things

  7. Nice summation of different competitive formats. I suggest picking the format that best suits the goals of the tournament. In the three armoured feats of arms I've participated in (and those I've helped organize), fighters were arranged as "appellants" and "defendants," with the appellants being the visitors and the defendants being the home team.

    The primary goals of the tournaments were to display skill and prowess, although two of them were "scored" by bouts won and lost for each side. Obviously, this isn't a competitive format of the types described above, but it is a good way to display skill at arms.

    Sean
    Last edited by Sean Hayes; 12-21-2010 at 10:42 AM.
    Sean Hayes, Maestro d'armi
    Northwest Fencing Academy

    Chivalric Fighting Arts Association
    San Jose Fencing Masters Program Examination Board

    One should never confuse the rules of a competition with the rules of an art.

    People talk a lot about speed, but not very much about control, safety, tactics, and trying to get close to the reality of sharps. When simulating sharp fights, how fast one charges in depends on how quickly one would like to die.


Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •