New Rank Comparison List (mid 2010):
Since the Worldwide Rank Comparison List seems to be incorrect in many cases, I have created a new list that rather specifies ranges of ranks instead of exact ranks. Unfortunately I have no information about ranks in China, Korea and Japan, but at least for Japan I know that there seems to be a high bandwith. While one of my friends is 5k in Japan, he's 2-3k on KGS. Another friend who was 6k EGF played even with Japanese 1d players.
The new comparison list that I have created is based on various sources (total number of datasets in brackets):
Number of information for each server/organization
Generally the specified ranges mean that they include half of the conversions from the given KGS rank (25% each above and below the median), hence you would have a 50% chance to fit within the range of another server/organization relative to your KGS rank. The bandwidth of the remaining 50% is much bigger. For some ranks of some servers only few data exist, so in case of less than 5 datasets per rank-server-server combinations, several ranks have been combined and redistributed with the proposition that an increase of one rank on one server means an increase of one rank of the other server (which is only approximately true). The base of the list are KGS data, i.e. data for other servers/organizations are derived from KGS data. The intervals for the KGS columns have been recalculated from DGS and EGF columns.
JTron: Has anything changed in the ratings systems in the last six months? In other words, is this still accurate? I like this table better than the other chart, but I wish the ranges in rank were a little closer. Otherwise, this has been very helpful to me. Thanks.
sh: As far as I can judge the comparison still looks good, though I haven't followed the changes since mid 2010 in detail. I am sorry that I cannot help with the wide ranges, but this is the way it is ;-)