Ranking of go players

There are systems of ranks and classifications that measure skill in go. Traditionally, go rankings have been measured using a system of dan and kyu ranks. Especially at amateur levels, these ranges facilitate the handicap system, with a difference of one range roughly corresponding to one free move at the start of the game. This system is also commonly used in many East Asian martial arts, where it often corresponds to a belt color. With the ready availability of calculators and computers, classification systems have been introduced. In such systems, a rating is rigorously calculated based on the results of the game.

Kyu and dan ranks

Traditionally, the rankings of go players have been measured using a system of kyū (級）and dan段) (both meaning 'level' or 'grade').

In China, Japan and Korea, according to tradition, to classify the categories of professional and amateur players of go, there are two different rankings: the lower one (the kyū) and the upper one (the give).

The kyū and the amateur dan (the latter not present in China) are used to classify amateur players, while the dan professionals are used to classify players who play Go professionally and participate in professional tournaments.

The kyū range from 30 kyū (lowest) to 1 kyū (highest).

In dan, both amateur and professional, it is the opposite of kyū. The lowest level is 1 dan, up to 9 dan. Therefore, a player who is 1 kyū would become 1 dan if he continues his progression.

In China the open system is practiced: it does not distinguish between professionals and amateurs, using the same criteria to award the titles. Not so in Japan, whose academy treats professionals more rigorously.

In some European countries, such as Germany, the two categories have been unified into one, made up of 18 degrees in total.

(There is also an 8 dan amateur title and a 10 dan professional title, but these are not the same as ranks.)

Although almost all organizations use this system, there is no universal calibration. The means of awarding each of those ranks and the corresponding levels of strength vary from country to country and between go online servers. This means that a player who is considered 2 kyu in one country, may be considered a 5 kyu in another.

Origin

The first ranks of Go occurred in China in the II century, when Handan Chun (Chinese: 邯郸淳) he described the classification system of 9 Pin Zhi (九品制) in his book Classic of Arts (艺经). From the early 17th century century, the Japanese formalized the teaching and classification of Go. The system was later used in the martial arts schools; and is believed to be ultimately derived from China's nine-rank system. The fact that there are 9 degrees of professional dan is believed to be based on the original 9 degrees of Chinese Pin Zhi.

Differences in strength up to the level of amateur dan generally correspond to the level of handicap that would produce even play between the two players. For example, a 3d player is expected to be able to give 2 handicap stones to a 1d player and win half the game for him. In contrast, the differences in the professional ranges are much smaller, perhaps 1/4 to 1/3 of a handicap stone. There are some significant differences in strength between normal 9p players and the world's best players because 9p is the highest rank possible, which may explain this variance.

Reach a dan rank

The ranks of dan (abbreviated online as "d") are for advanced amateur players. Although many organizations allow players to choose their own kyu rank to a certain extent, dan ranks are often regulated. This means that players will need to show good results in tournaments or pass exams to get a dan rank. Serious students of the game will often strive to achieve a dan rank, just as practitioners of martial arts will strive to achieve a black belt. For hobbyists, ranks dan up to 7th dan are available. Above this level, a player must become a professional player to achieve further promotions. In Japan and China, some players are awarded an amateur eighth dan, rank as an honorary title for exceptional achievement. In the United States, amateur dan ranks are often based on the AGA ranking system. Under this system, some strong amateurs and former professional players have reached up to amateur 9th dan, although they will generally register as 6th or 7th dan in events. international. Similarly, some players have reached the 9th dan of amateurs in the online go server ranking system.

Although players who have reached professional dan ranks are nominally stronger than amateur dan players, in practice some of the 7th dan amateur players Stronger >dan have a level of play on par with some professional players. These players have never tried to reach professional rank or have chosen to remain amateur players because they do not want a career playing go.

Professional Ranks

The professional dan ranking system is similar to amateur ranks in that it awards dan ranks that increase numerically with skill. However, the difference between these grades is much smaller than with amateurs and is not based on the number of handicap stones required. Professional dan ranks go up to 9th dan, but the difference in strength between a 1st dan dan and a professional 9th dan generally does not exceed a 2-3 stone handicap.

To distinguish between professional dan and amateur dan, the former is often abbreviated "p" (sometimes called ping) and the second "d". No such abbreviation existed in the past, and it is not generally used as an abbreviation beyond the Internet, where it is common, but not universal.

Classification system

With the invention of calculators and computers, it has become easy to calculate a ranking for players based on the results of their games. Commonly used rating systems include the Elo and Glicko rating systems. Ranking systems generally predict the probability of one player defeating another, and use this prediction to rank a player's strength.

Elo rating system used in go

The implementation of the EGF (European Go Federation) Elo system attempts to establish a rough correspondence between ratings and kyu/dan ranks. This is done by varying some of the components of the Elo formula to achieve a close match to the adjacent table. The probability (S_E) that the lowest ranked player, Player A, will win against the highest ranked Player B is given by the formula:

SE(A)=1eD/a+1{displaystyle S_{E}(A)={frac {1}{e^{D/a}1}}}}}

• D is the score difference: RB− − RA{displaystyle R_{B}-R_{A},}
• a varies according to player A's previous rating

The probability that player B wins is calculated as:

SE(B)=1− − SE(A){displaystyle S_{E}(B)=1-S_{E}(A),}

A player's new rating is calculated as:

Rn=Ror+K(S− − SE){displaystyle R_{n}=R_{o}+K(S-S_{E}),}

• R_n = new score
• R_or = old score
• S = result (1, 0.5 or 0)
• S_E = expected result
• K varies depending on the player rating

K varies by player rating, due to low confidence in (lower) amateur ratings (high fluctuation in result) but high confidence in professional ratings (stable and consistent play). K is 116 with a rank of 100 and 10 with a rank of 2700

In the EGF system, the Elo points gained by the winner are almost equal to those lost by the loser and the maximum movement of points is the constant K (from above). However, there is a slight inflationary mechanism built into the ratings adjustment after each game to compensate for the fact that newcomers generally bring less ELO points to the pool than they bring with them when they stop actively playing. Other Elo-like ratings such as the AGA, IGS, and DGS systems use maximum likelihood estimation to fit ratings, so those systems are anchored by priors rather than trying to ensure that the gain/loss of ratings is summative. zero.

Other classification systems

A variation of the Elo system called Whole History Rating (WHR), differs from standard Elo in that it retroactively re-ranks players based on their full history after each new result is added, instead of gradually changing the ranking game by game. This is more computationally intensive than other methods, but it is claimed that "compared to Elo, Glicko, TrueSkill and decaying history algorithms, WHR produces better predictions". The Go Ratings website implements the WHR method to calculate the global ranking of the players.

Rating Base

Player rankings are generally measured using the game results of go tournaments and competitions. Most clubs and countries maintain their own rankings, as do the go servers. In Europe, the official EGF classifications are used.

In a small club, ranks can be decided informally and manually adjusted when players consistently win or lose. In larger clubs or country-wide ranking systems, a mathematical ranking system is generally easier to maintain. Players can be promoted or demoted based on their strength calculated from their wins and losses.

Most go servers use a mathematical ranking system to keep track of the playing strength of their members. Such rankings may or may not be translated into kyu and dan ranks for the convenience of players.

Player groups that don't mix regularly (such as different countries or subgroups on online servers) often result in diverging play strengths compared to the same nominal rank level of other groups. Therefore, players who are asked to give their rank often rate it with "in my country" or "on this Internet server".

Odds of Winning

The ranking indirectly represents the probability of winning an even game against other qualified players. This probability depends only on the difference between the two players' ratings, but its magnitude varies greatly from implementation to implementation. The American Go Association adopted a uniform standard deviation of 104, that is, slightly more than one range, while the European Go Federation scores have a standard of deviation ranging from 200 for beginners to 70 for the best players. The IGS has fixed a standard deviation for all levels, but with a non-standard distribution. The following table shows some of the differences:

Probability of victory and handicaps

Whereas in chess a player must take some risks to avoid a tie, in go ties (jigo) are impossible (with non-integer superko and komi, like 6.5 points, as is common). or less likely in the case of integer komi. Also, an average game of go is 240 moves long (120 moves in chess terms), compared to 40 in chess, so there are more opportunities for a weaker player to make sub-optimal moves. The ability to turn a small advantage into a win increases with playing strength. Due to this ability, stronger players are more consistent in their results against weaker players and will generally earn a higher win percentage against opponents at the same range distance.