Page 5 of 5 FirstFirst ... 345
Results 61 to 66 of 66
  1. #61
    Monster of the deep Kevin Bonham's Avatar
    Join Date
    Jan 2004
    Posts
    39,807
    Quote Originally Posted by Patrick Byrom View Post
    Because it uses players from the same year, it also avoids another concern Kevin raised - which I share - that the difference may not have been constant over the past forty years.
    Actually there would be a lot of merit in using many different years as possible re the Aus Junior. For instance suppose that some year some state has an unusually good coaching or assistance program that causes its players to overperform compared to ratings in that year specifically.

    I was interested with the Aus Junior in seeing if there was an effect across the range of age groups and in both a Queensland and a non-Queensland event, and it looks like there is, but it's a small part of the puzzle.
    Moderation Requests: All requests for, comments about, or questions about moderation of any kind including thread changes must be posted in the Help and Feedback section and not on the thread in question. (Or by private message for routine changes or sensitive matters.)

    ACF Newsletter Information - All Australian players and administrators should subscribe and check each issue for relevant notices

    My psephology/politics site (token chess references only) : http://kevinbonham.blogspot.com.au/ Politics twitter feed https://twitter.com/kevinbonham

  2. #62
    CC Candidate Master
    Join Date
    Jun 2009
    Posts
    41
    Quote Originally Posted by Patrick Byrom View Post
    That's an excellent idea, but I'm not sure there is enough data to obtain reliable results if it is broken into groups.
    My idea was to see how much of the variation is explained by these other factors (as opposed to the state), rather than to split the sample into groups. Essentially, regressions. You'd still be using (almost) the entire sample in the analysis, but you would be closer to comparing apples to apples.

    It's just another idea - I agree that a combination of approaches is useful, including what you and Kevin have already proposed.

  3. #63
    CC FIDE Master
    Join Date
    Jan 2004
    Posts
    714
    Quote Originally Posted by Kevin Bonham View Post
    Actually there would be a lot of merit in using many different years as possible re the Aus Junior. For instance suppose that some year some state has an unusually good coaching or assistance program that causes its players to overperform compared to ratings in that year specifically.

    I was interested with the Aus Junior in seeing if there was an effect across the range of age groups and in both a Queensland and a non-Queensland event, and it looks like there is, but it's a small part of the puzzle.
    I think the Aus Junior comparison would be extremely interesting as the dynamic has changed significantly over the years from the old 1 round a day format, through to a much more condensed event in it's current format. I remember the old format involving a lot more preparation for games as coaches had more time to be able to do this at all age levels and the fact the event was more spread out allowed for regular bulletins with games to be produced. Those states that handle different formats better based around their approach could definitely end up skewing results to a degree. The other thing to factor in is that if you get a group of kids around the same age pushing each other, you often develop a really strong group within a state that lasts however many years - as evidenced by the Queensland cohort in the Under 12 and below category currently, and as another example the group of NSW kids that went through which contained Kuan Kuan Tian, Michael Lip, etc or the Victorian group that came out of Melbourne High and Scotch roughly 5-6 years ago.

  4. #64
    CC Grandmaster
    Join Date
    Apr 2008
    Posts
    6,770
    Quote Originally Posted by smurfo View Post
    My idea was to see how much of the variation is explained by these other factors (as opposed to the state), rather than to split the sample into groups. Essentially, regressions. You'd still be using (almost) the entire sample in the analysis, but you would be closer to comparing apples to apples. It's just another idea - I agree that a combination of approaches is useful, including what you and Kevin have already proposed.
    I think I understand what you're suggesting now (using multiple regression) but I can't see how to implement it here when there is no obvious independent variable. But I'm probably missing something obvious

    What I did do was to split the state data into different age groups. And then compare the differences in the average rating between states for players born in each decade from 1960s to the 2000s. Qld average ratings were consistently below those of NSW and Victoria. However, as you'd expect, there was some variation between the decades, with the smallest difference being about 50 and the largest over 300. I didn't examine the distributions, so these may be skewed in some cases.

    An interesting result was that the average rating steadily increased as the players got older - except for Queenslanders born in the 1970s (I blame Sir Joh!). It's not too surprising that older players would generally be stronger, of course, but I suspect that there may other factors as well. The other (mildly) interesting result was that the large group of players in each state with no recorded date-of-birth seem to have similar average ratings to the players whose dobs have been recorded.

  5. #65
    CC Grandmaster
    Join Date
    Apr 2006
    Location
    Melbourne, Australia
    Posts
    15,786
    Smurfo is absolutely right! Got to be a multi-dimensional analysis.
    So juniors of QLD got to be compared with Juniors of other states etc.

    Re Queenslander ratings being consistently below than those of NSW and VIC: Historically (correct me if I am wrong) level of chess in QLD has been lower than in Vic or NSW. If we look at the cross-tables for national champs from the 1950s-80's? How many QLD players have been performing well? Solo (Qld's best player for many years till the rise of David S. occurred) is originally from Melbourne.
    Interested in Chess Lessons?
    Email webbaron!@gmail.com for more Info!

  6. #66
    CC Grandmaster
    Join Date
    Apr 2008
    Posts
    6,770
    Quote Originally Posted by MichaelBaron View Post
    Smurfo is absolutely right! Got to be a multi-dimensional analysis. So juniors of QLD got to be compared with Juniors of other states etc.
    Which I just did.

    Quote Originally Posted by MichaelBaron View Post
    Re Queenslander ratings being consistently below than those of NSW and VIC: Historically (correct me if I am wrong) level of chess in QLD has been lower than in Vic or NSW. If we look at the cross-tables for national champs from the 1950s-80's? How many QLD players have been performing well? Solo (Qld's best player for many years till the rise of David S. occurred) is originally from Melbourne.
    The national rating system in its present form only dates back to 1980, so comparisons before then wouldn't tell us much about rating differences. And I don't believe that there is a significant difference in ratings between states for players rated over 2 000 anyway, as they play each other more regularly.

Thread Information

Users Browsing this Thread

There are currently 1 users browsing this thread. (0 members and 1 guests)

Similar Threads

  1. Replies: 24
    Last Post: 22-08-2016, 03:28 PM
  2. Income distribution
    By Johns in forum Politics
    Replies: 3
    Last Post: 17-07-2015, 11:35 AM
  3. prize distribution
    By FM_Bill in forum Arbiters' Corner
    Replies: 9
    Last Post: 24-08-2013, 06:22 PM

Posting Permissions

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