Introduction

The goal of the rating tournament is to allow for the issuing of ratings certificates up to 1D, based upon the results of a 5 game tournament.

This approach to issuing children’s certificates is very common in China. As schools have started opening in Canada with a significant Chinese background, they have been looking to offer their students something similar here in Canada. The first school to offer such certificates in collaboration with the CGA was Golden Key, starting around 2014. Then in 2017/2018 Gofun Studio and We-Go schools joined in, and in 2019 Ottawa Go school will join the program.

Although Chinese schools issue such certificates, the meaning of Kyu ranks is very different from North America/Europe. As an example look at the following table:

Recent rank comparison

You can see a 15K in North America/Europe would likely be 4K in China.

What are our goals when issuing these certificates?

We want any children who have made noticeable progress since their last rating tournament (or if this is their first event) to be very likely to earn a certificate. This can reassure parents that their children are making progress, and encourage the children to progress further.
Especially as we go above 10K we would like the ranks to roughly align with “normal” CGA rated kyu ranks. At this level the children may also participate in regular adult tournaments, where they will encounter adults with a similar rank. The children will expect to be able to enter at their certificate rank and be competitive with these adults. We don’t do them any favors by granting them inflated rank certificates, and then when they attend an adult event they mostly lose and become demoralized.

To avoid problems with either inflated ranks, or too many children stuck without earning a certificate we want to ensure we issue more finely grained certificates. I.E instead of just issuing a 20K and 15K certificate we should provide a path to earn 19K, 18K, etc.

Handicap details

The novice group (21K – 40K)

Our “mental model” assumes an adult novice (probably with some general gaming experience) will after a handful of 9×9 games be 25K. The basic standard for 25K is to be able to functionally complete a Go game. They should roughly know how to score, and the rules. A child novice especially near 5 years old could start out much weaker than this, but we don’t want to track progress below this standard. Children attending a few months of classes would normally reach this level.

Children entering the rating tournament with a rating below 20K should compete in the novice group. If they can win 1 game they can earn a 25K certificate. 2 wins would go to 23K (or if they entered at 23K they could move up to 21K). 3 win games will move them to 20K. If two novice’s are paired they can play on the 13x13x board, at the organizers discretion. Also at the organizers discretion the novice group can be run completely separate from the rest of the ratings tournament, and optionally fully on 13×13 (since 13×13 games may be much faster). A “bye” round for any child should count as a win. 4 wins can move them to 19K, and 18K for 5 wins.

If novices are paired up with 20K+ group they can use a 3 stone handicap.

Rank differences and handicaps

Some context here, to explain the thinking in in the advancing groups. At any level, our expectation is (if the ranks are correct) handicap cap of rank A – rank B should give a 50% win chance. I.E a 7D should have a 50% win chance against a 5D with two handicap. And an 18K should have a 50% win chance against a 16K with two handicap. However there are two factors that cause us to use lower handicaps at the high Kyu ranks:

We don’t have great confidence the 18K & 16K ranks are correct.
The result is much more random at the 18K level. For example if the 5D/7D played without handicap the 7D might win 95%. At 16K/18K it might be only 60/70%.

The 20K – 16K group

At this level we now expect all games to be played 19×19. Here we will play handicap = rank difference -2. So

19K versus 20/18/17 would all be no handicap
20K versus 17K would be “No Komi” (white wins a tie on the board, or 181 stones to win Chinese counting)
20K versus 16K would be 2 stones

Then for promotion in this group:

To get to 15K(promote to the next group) you must score 3 wins. If for example you were 16K and won two games you cannot move up. If you do score 3 wins you can move the number of wins. For example a 16K winning 3 can move to 13K. A 16K scoring 4 could move to 12K. If you are 20K/19K and score 3 you would move to 15K.
Within the group you may move up two ranks for scoring two wins. I.E a 19K can move to 17K with 2 wins.

If paired up use the handicap rules from the 20K – 16K group (rank difference – 2).

The 15K – 11K group

The rules here are the same as the 20K-16K group (handicap of rank difference – 2). Except here if paired up we use the handicap rules from the group we pair up to. From 10K+ we need to make sure they are meeting a tougher standard to promote, so they must pair using a tougher handicap standard.

As a 15K – 11K you need 3+ wins to promote to 10K, as a 15-13K you could promote two ranks with two wins. 4 wins earns 9K, 5 wins earns 8K.

The 10K – 6K group

Here we will play handicap = rank difference -1. So

10K versus 9K would be no handicap
10K versus 8K would be “No Komi” (white wins a tie on the board, or 181 stones to win Chinese counting)
10K versus 6K would be 3 stones
10K versus 5K would be 5 stones

If you score 3 wins you move up one rank. 4 wins two ranks, and 5 wins 3 ranks.

The 5K – 1K group

Here we play full handicap (including if paired up again the 6K- 10K group). Up to 1K you can promote one stone with 3 wins, two stones with 4 wins, and 3 stones with 5 wins. But promotion to 1D is special, it doesn’t work like this. You can only get to 1K with the rules above. To get to 1D you must enter the rating tournament as at least 2K. Then

As a 2K you must score 5 wins to become 1D
As a 1K 4 wins moves you to 1D.

Additional background context

Normally at the heart of a rating system is many players who are fairly stable. But for children’s promotion this is not the case. The majority of the children are actively studying, and the majority may have changed level since their last tournament. This can cause problems in a couple ways:

If the children are playing only one rating tournament a year, it is possible this system doesn’t promote fast enough.
In the Toronto area especially, there may be several ratings tournaments. If a group of children attend several of these, over the multiple events they may move up more than appropriate. And by playing within the same group they can all “move up together”.

Generally we depend on the Go schools/teachers to monitor for these issues. They can bump up children at the entry to the next rating tournament they participate in. If it seems promotion is happening too fast we could adjust with a more finely grained promotion (especially in the 20K-10K band). Our general guidance when entering children into the system for the first time (outside the novice group) would be “be conservative in choosing their entry rank”. This way the children can have a good chance to earn a certificate.

Also, if children enter outside the novice group from 20K -11K with 1 win they can be granted the rank they entered at. From 10K+ they need two wins to be granted their entry rank. 1 win they can receive two stones below their entry rank. I.E if they entered this system for the 1st time as an 8K and earned 1 win they can get 10K certificate.

Initial notes on Dan promotion

The general principal is to earn a 2D+ Dan certificate they need to earn a “stable CGA rating” at the rank they wish to earn a certificate for.

This would mean

If they promoted from below, they should get up to 1/4 of the rank. I.E if exiting the children’s system they were ranked at 50. They should get to a CGA numerical rating of 125. (100-199 is 2D). We don’t want them to get “just over” 2D, get their certificate, and then fall back in rating to 1D
Some children will come in (either from abroad, or from a part of the country where there isn’t a rating tournament). They might be entered at 3D (200-299) initially. This would give them an initial score of 250. But they would need to have played at least 10 games and keep their score above 225 to earn the certificate. 10 games would be 3 adult one day tournaments.

Many children’s tournament’s you may want to include Dan players. This is fine/encouraged, you would have a “champion’s group” as part of the event. But usually you would do the following things:

Award trophies, since normally the Dan children cannot earn rating certificates.
Submit the dan players results to the CGA ratings.
Play at their games at full handicap. You should explain that this gives the chance for both players in the game to earn significant “CGA ratings points”.
Check with the CGA post event if any of the children meet the “Dan promotion criteria”.

2019/2020 season fees

25K – 6K : 10$
5K – 1K : 15$
1D+ : 20$