-3

Re the close: If it was really a duplicate then why didn't Brian Towers♦ answer on the other post? Checkmate!


On Wiki, we have the List of youngest grandmasters since 1950. What about WCMs (woman candidate masters) instead of GMs?


follow-up questions to brian towers' answer in case they get deleted (I notice that on this stackexchange site, follow up questions or reasons for downvoting can be removed without warning or explanation!):

  1. what is this database please? may you please share the database or how it was constructed or something here? chess.stackexchange.com/a/37013 i mean: did you really download all the stuff since 2001 and then combine all the data into 1 huge database? or what? or did you get it from somewhere or what?

  2. as for re research effort: it seems the only way to do this is to get the database. so how? really downloading and combining everything? seems pretty long to do. i figured somehow there was something already constructed. if i do this from scratch, then i risk reinventing the wheel. am i wrong?

  3. do we have the same 'procedure/algorithm' here chess.stackexchange.com/questions/36988/


Youngest CMs: List of records for youngest CMs (candidate masters) since 1950 (or 2014 or whatever)?

Youngest female CMs: List of records for youngest female CMs (candidate masters) since 1950 (or 2014 or whenever)?

Youngest WCMs: List of records for youngest WCMs (woman candidate masters) since 1950?

1
  • 2
    It would be great if you could collect the links to all of your dozens of questions in this format rather than sporadically choosing 2-3 to reference in each new post Oct 9 at 21:20
4
+50

Perhaps worth noting why the question has attracted down votes. If you hover your mouse over the down arrow it starts off by saying :This question does not show any research effort".

It is worth pointing out that one of the best places to start your research is this forum. Searching for questions about candidate master titles gives a number of useful hits.

This question - When was the FIDE title of Candidate Master first awarded? - has the very useful answer -

The first rating list that contains the Candidate Master title is April 2002 and all the players on that list with the said title are from England followed by Wales in the next rating list (July 2002)

The CM and WCM titles were created at the same time and so there were no WCM titles anytime before 2002 and certainly not in 1950.

Furthermore the CM and WCM titles are not particularly prestigious. Amongst other ways of getting these titles is by finishing in the first 3 in continental championships for children as young as 7 or 8. Try doing an advanced search on the FIDE rating website for WCM players and you will see players with ratings just above the 1000 threshold. I think this change to the title requirements was made in 2012 so players awarded the title before that were 2000 rated or over.

Searching my database constructed from data available on the FIDE Ratings Download page gives the following results. Note that FIDE gives only the birth year for players due to data privacy laws so age was calculated as 'Year of title' - 'Year of birth' so some of the players marked as 8 years old may have actually been only 7 years old when they won the title.

Year Name Age Fed
2002 Haslinger, Mandy S 22 ENG
2003 Wilson, Alexandra 17 ENG
2004 Liu, Yang Hazel 14 SGP
2005 Nguyen, Thi Mai Hung 11 VIE
2005 Phan, Nguyen Mai Chi 11 VIE
2008 Chiang, Sarah 11 USA
2008 Kiseleva, Arina 11 RUS
2009 Ivana Maria, Furtado 10 IND
2012 Salonika, Saina 9 IND
2012 Solozhenkina, Elizaveta 9 RUS
2012 Tran, Vuong Mai Khanh 9 VIE
2013 Ismayil, Malak 9 AZE
2013 Pham, Tran Gia Thu 9 VIE
2019 Bashirli, Saadat 8 AZE
2019 Zhumagali, Raian 8 KAZ

10
  • thanks Brian Towers♦! thing1: what is this database please? may you please share the database or how it was constructed or something here? chess.stackexchange.com/a/37013 i mean: did you really download all the stuff since 2001 and then combine all the data into 1 huge database? or what? or did you get it from somewhere or what?
    – BCLC
    Sep 18 at 23:23
  • thing 2 as for re research effort: it seems the only way to do this is to get the database. so how? really downloading and combining everything? seems pretty long to do. i figured somehow there was something already constructed. if i do this from scratch, then i risk reinventing the wheel. am i wrong?
    – BCLC
    Sep 18 at 23:26
  • 2
    @BCLC you can either implore everyone who might have constructed this database (look around) to send it over to you / open-source it, or assume they won't be willing to do so and construct it yourself. What is definitely a waste of time is any other approach (like trying to collect piecemeal factoids about these records). Sep 26 at 1:43
  • 1
    @MobeusZoom My database comes in at about 15 GB. It's not something I can email to you and it will cost me money to publish it. Add to that the enormous amount of time and effort (mostly in cleaning early data which was very dirty) in generating it in the first place and you have to have a screw loose to think anybody is going to do that. See also this answer for a glimpse into the world of dirty FIDE rating data - chess.stackexchange.com/questions/10975/…
    – Brian Towers
    Sep 26 at 9:58
  • 1
    @BrianTowers I'm not asking for it of course, just suggesting that it's a necessary approach for something like BCLC's questions. With that said, in my opinion the more work it took you - I can see it was a great amount, which only adds to your contribution - the more worthwhile it is to open-source such a resource. (I believe that Kaggle would host it for free at 15 GB but there may be other better places. I'm not an expert.) I don't think "you have to have a screw loose to think anybody is going to do that" - open-sourcing datasets that required tremendous work is how scientists contribute. Sep 26 at 17:51
4
+100

what is this database please?

The database I use is an SQL Server database I built several years ago using data downloaded from Olimpbase for the years 1967-2001 and the FIDE download page for 2001 to today. Note that while the Olimpbase link I have given is for one zip file containing all the data 1967 to 2001. The FIDE page, however, requires you to select and download each period individually. In the early years after 2001 that was 4 files per year. Later that moved up until for the last few years it is one per month, 12 per year.

This is a massive amount of very dirty data. The current download for September 2021 holds data for 1,025,666 players. My current database comes in at over 15 GB. The idea that you could download it from me is nonsensical.

Regarding research, you have obviously done none. The obvious place to start is this community. Had you done so you would have likely found this similar question - Creating program for finding top players with respect to elo and age - and with it my answer. The final paragraph of my answer is particularly relevant -

A major task will be extracting the data you need from your data sources. This will be a massive task. I would suggest you design a relational database for your needs and populate it from your data sources after first cleaning the data. Once this is done answering the questions you want is a relatively straightforward matter of writing some simple database queries and running them.

In summary:

  1. Download all the data
  2. Identify "dirty data" and clean it
  3. Design your database bearing in mind the structure of the raw data and how you want to use that.
  4. Load the data into your database.

There is a fifth, ongoing task:

  1. Every month download the latest data from FIDE, check it for dirty data (unfortunately there is still some despite my regular complaints to the relevant FIDE officer when I get fed up), clean the dirty data and load it into the database.

On the dirty data, front check out this answer I gave to an earlier question relating how FIDE and the federations abuse FINS. I still haven't gotten over what they did to Tony Miles.

1

Not the answer you're looking for? Browse other questions tagged or ask your own question.