Page 6 of 6 FirstFirst ... 456
Results 76 to 90 of 90
  1. #76
    Reader in Slood Dynamics Rincewind's Avatar
    Join Date
    Jan 2004
    Location
    The multiverse
    Posts
    21,567
    Yes, well done Neil Wright, especially after Kevin Sheldrick's hot start.
    So einfach wie möglich, aber nicht einfacher - Albert Einstein

  2. #77
    CC Grandmaster Garvinator's Avatar
    Join Date
    Jan 2004
    Location
    Brisbane
    Posts
    13,073
    Quote Originally Posted by Kevin Bonham
    Neil also won in 2001-2 - winning this event twice out of three consecutive holdings is a very impressive feat for any player.
    I believe that the winner of the major gets automatic entry to the championships, so i take it that Neil just decided not to play in the championship in Adelaide, which explains his absense.

  3. #78
    CC Grandmaster
    Join Date
    Jun 2004
    Posts
    5,667
    Quote Originally Posted by ggrayggray
    I believe that the winner of the major gets automatic entry to the championships, so i take it that Neil just decided not to play in the championship in Adelaide, which explains his absense.
    Is he allowed to play the Major (assuming his rating is below 2150)?

  4. #79
    Illuminati Bill Gletsos's Avatar
    Join Date
    Jan 2004
    Location
    Sydney
    Posts
    16,544
    Quote Originally Posted by ggrayggray
    I believe that the winner of the major gets automatic entry to the championships, so i take it that Neil just decided not to play in the championship in Adelaide, which explains his absense.
    Neil played in the 2003-2004 Australian Championship in Adelaide.
    The Force can have a strong influence on the weak-minded.

  5. #80
    Illuminati Bill Gletsos's Avatar
    Join Date
    Jan 2004
    Location
    Sydney
    Posts
    16,544
    Quote Originally Posted by pax
    Is he allowed to play the Major (assuming his rating is below 2150)?
    Yes.
    The Force can have a strong influence on the weak-minded.

  6. #81
    Illuminati Bill Gletsos's Avatar
    Join Date
    Jan 2004
    Location
    Sydney
    Posts
    16,544
    Using ACF ratings here are the rating performances as calculated by Swiss Perfect.
    Note where the player had no ACF rating their FIDE rating was used.
    Code:
    No  Name                       Feder Loc Id  Loc  Score  Exp.  Rav Rprfm
    
    1.  Sheldrick, Kevin           SA    5201261 2118 7.5/11 8.14 1932  2065
    2.  Norris, Damian             QLD   4137082 2084 7.0/10 8.30 1816  1965
    3.  Tredinnick, Malcolm        NSW   2090484 2073 8.0/11 8.69 1840  2015
    4.  Jones, Lee                 NSW   2044155 2054 5.5/10 7.20 1890  1926
    5.  Duxbury, Craig             QLD   4141153 2045 3.5/5  3.55 1883  2032
    6.  Milligan, Helen            SCO           2045 6.0/11 8.36 1842  1878
    7.  Marner, Gavin              NZL           2044 6.5/11 7.81 1884  1949
    8.  Hu, Jason                  NSW   2069944 2024 6.0/11 8.91 1778  1814
    9.  Dizdarevic, Mehmedalija    VIC   3101894 2014 6.5/11 7.15 1904  1969
    10. Sonter, Matthew            QLD   4157611 2013 6.5/11 8.58 1793  1858
    11. Zvedeniouk, Ilia           NSW   2204212 2001 8.0/11 6.82 1913  2088
    12. Viner, Phillip J           NSW   2032244 1971 5.5/10 7.70 1764  1800
    13. Wright, Neil               NSW   2049145 1945 8.5/11 6.16 1901  2112
    14. Mendes da Costa, Alex      NSW   2073240 1930 6.5/11 7.48 1794  1859
    15. Krstic, Slobodan           VIC   3090742 1928 7.0/11 6.49 1863  1965
    16. Fry, Peter                 VIC   3074911 1908 5.5/10 6.80 1773  1809
    17. Stojic, Svetozar           VIC   1281340 1879 7.5/11 5.83 1859  1992
    18. Truscott, Tony J           QLD   4139764 1879 7.5/11 4.95 1915  2048
    19. Davidovici, Victor         QLD   4132322 1861 4.0/10 6.70 1737  1665
    20. Lilly, Richard             WA    6232956 1851 4.0/11 8.25 1656  1554
    21. Vijayakumar, Rukman        VIC   3108821 1851 5.0/9  3.24 1950  1993
    22. Rout, Ian                  ACT   1272020 1846 4.0/10 7.30 1674  1602
    23. Cashman, Michael           QLD   4131644 1842 7.0/11 4.84 1888  1990
    24. Myers, John                QLD   4138894 1833 5.5/11 6.49 1766  1766
    25. Tulevski, Vasil            NSW   2031555 1828 4.5/10 5.50 1795  1759
    26. Davis, Tony                VIC   3073264 1816 5.5/11 5.94 1790  1790
    27. Lukursky, Boris            NSW   2022525 1784 4.5/11 8.25 1593  1528
    28. Lazarus, Benjamin          QLD   4163344 1771 8.0/11 4.29 1850  2025
    29. Selnes, Hamish             QLD   4138740 1752 4.5/11 5.94 1722  1657
    30. Al Zaher, Louay            QLD   4198333 1737 4.0/9  5.22 1680  1637
    31. Barisic, Frank             NSW   2011056 1731 3.0/9  6.12 1599  1474
    32. Lovejoy, David             QLD   4161841 1730 6.5/11 3.96 1836  1901
    33. McKenzie, Colin            VIC   3080485 1716 4.0/10 5.30 1694  1622
    34. Holland, Dennis            WA    6249270 1715 5.0/10 5.20 1704  1704
    35. Hackenschmidt-Uecker, Jorg QLD   4536462 1708 5.5/11 5.72 1691  1691
    36. Guo-Yuthok, Sherab         ACT   1279703 1706 5.0/10 5.30 1684  1684
    37. Cox, Barry                 NSW   2067251 1701 6.0/11 3.19 1862  1898
    38. Hvistendahl, Robert        NSW   4186215 1701 5.0/9  2.25 1898  1941
    39. Oliver, Shannon            ACT   1276956 1685 6.0/11 4.73 1733  1769
    40. Korenevski, Oleg           QLD   4163543 1674 5.5/11 4.18 1759  1759
    41. Myers, Stephen             NSW   3081605 1661 4.5/10 5.30 1639  1603
    42. Canfell, Mike              NSW   2013252 1648 5.0/10 3.80 1736  1736
    43. Brockman, Roland           VIC   3071676 1627 4.0/10 2.60 1813  1741
    44. Schon, Eugene              VIC   3115450 1611 5.0/11 3.19 1767  1731
    45. Behne-Smith, David         NSW   2067833 1585 5.5/11 2.75 1776  1776
    46. Humphries, Ryan            WA    6246470 1555 3.0/10 5.50 1519  1370
    47. Barker, Ken                QLD   4130546 1526 5.0/11 6.38 1466  1430
    48. Chuang, Howard             QLD   4200945 1514 5.5/11 3.19 1675  1675
    49. Tang, Jason                VIC   3116953 1465 4.5/11 6.60 1390  1325
    50. Tangimentua, Tyson         QLD   4182940 1360 5.5/10 4.40 1400  1436
    51. Horton, Russell            TAS   7295405 1357 4.0/11 3.30 1503  1401
    52. Russell, Luthien           QLD   4206372 1334 4.5/9  2.88 1469  1469
    53. Slack-Smith, Blair         WA    6249163 1290 4.5/11 1.87 1566  1501
    54. Hunter, Shayne             QLD   4530265 1235 3.0/10 2.20 1456  1307
    55. Guo, Emma                  ACT   2070736 1197 3.5/10 2.90 1351  1241
    56. Bhattacharya, Devraj       VIC   3121444 1112 3.0/10 1.70 1387  1238
    57. Bielenberg, Nathanael      QLD   4531772 1112 4.0/11 1.43 1432  1330
    58. Long Hong, Stan            QLD   4135833 1094 1.0/10 2.30 1308   942
    59. Sheng, Susan               VIC   3120283 1085 0.5/9  1.80 1326   882
    60. Behne-Smith, Jonathan      NSW   1291583 974  4.5/10 1.40 1279  1243
    61. Feria, Antolin             QLD                4.5/11      1704  1639
    62. Soo, Brayden               QLD   4191930      6.0/11      1748  1761
    Last edited by Bill Gletsos; 14-01-2006 at 05:27 PM. Reason: removed rating gain as figure is misleading as it bears no relevance to the ACF system
    The Force can have a strong influence on the weak-minded.

  7. #82
    Monster of the deep Kevin Bonham's Avatar
    Join Date
    Jan 2004
    Posts
    37,479
    The top ten seeds all performed below their ratings, some badly so. Not surprised in some cases as FIDE-ACF parity doesn't exist at that low level and some of the Australian players' ratings have been on the skids for ages, but surprised to see Jason Hu have such a shocker.

    Excellent result for Rincewind - he got +1=6-2 against nine higher rated players, some of them over 300 points higher. (I'm especially respectful of this effort having just lost to Jones when I played him and been pretty badly walloped by Hu at Mt Buller.)

  8. #83
    Reader in Slood Dynamics Rincewind's Avatar
    Join Date
    Jan 2004
    Location
    The multiverse
    Posts
    21,567
    Quote Originally Posted by Kevin Bonham
    The top ten seeds all performed below their ratings, some badly so. Not surprised in some cases as FIDE-ACF parity doesn't exist at that low level and some of the Australian players' ratings have been on the skids for ages, but surprised to see Jason Hu have such a shocker.
    Jason was sick.

    Quote Originally Posted by Kevin Bonham
    Excellent result for Rincewind - he got +1=6-2 against nine higher rated players, some of them over 300 points higher. (I'm especially respectful of this effort having just lost to Jones when I played him and been pretty badly walloped by Hu at Mt Buller.)
    Actually got +2=6-2 vs 10 higher rated opponents However that should be tempered with the fact that 6 month ago my rating was 150 points higher than it is today. That has me scoring only 50 points over expectation. Still I was very happy with how I went.

    I think I caught Lee by surprise a bit and used a lot of time in the opening/middle which paid off, winning a pawn, but ultimately inhibited me from finishing off that game, and as I mentioned Jason was not well. I was also very lucky, scamming a win from Al Zaher and draw from Davis, both from lost positions.
    So einfach wie möglich, aber nicht einfacher - Albert Einstein

  9. #84
    Account Permanently Banned firegoat7's Avatar
    Join Date
    Jan 2004
    Location
    MCC
    Posts
    2,809
    Hello,

    What I find interesting is that more rating points are lost then gained out of the overall Australian rating pool, for this specific tournament, by this specific field. Is that normal? Is it anything to be worried about? Does it mean the Australian rating system is deflationary or is this a specific blimp? Do the unrated players count? What are the mathematical explanations in laymen terms?

    Just curious.

    I am not sure that it matters if it should be a zero total outcome, but you would think it would matter long term.

    cheers Fg7

  10. #85
    Reader in Slood Dynamics Rincewind's Avatar
    Join Date
    Jan 2004
    Location
    The multiverse
    Posts
    21,567
    Quote Originally Posted by firegoat7
    What I find interesting is that more rating points are lost then gained out of the overall Australian rating pool, for this specific tournament, by this specific field.
    Bill should confirm but my understanding is that the figures posted above are straight out of Swiss Perfect and the performance and rating change calculations are done according to FIDE's Elo system. Therefore they have no relevance to the Australian rating pool other than as a very rough guideline.
    So einfach wie möglich, aber nicht einfacher - Albert Einstein

  11. #86
    Illuminati Bill Gletsos's Avatar
    Join Date
    Jan 2004
    Location
    Sydney
    Posts
    16,544
    Quote Originally Posted by firegoat7
    Hello,

    What I find interesting is that more rating points are lost then gained out of the overall Australian rating pool, for this specific tournament, by this specific field. Is that normal? Is it anything to be worried about? Does it mean the Australian rating system is deflationary or is this a specific blimp? Do the unrated players count? What are the mathematical explanations in laymen terms?
    Those estimated rating point changes have nothing to do with the ACF rating system. They are estimates by Swiss Perfect based on the Elo system with a K factor of 10. Swiss Perfect is using a 350 point rating cutoff to determine the average rating of a players opponents which in and of itself is imprecise. I suspect this is the cause.
    They really should be disregarded. I really posted them to show the estimated performance ratings.

    Quote Originally Posted by firegoat7
    I am not sure that it matters if it should be a zero total outcome, but you would think it would matter long term.
    In correctly calculated Elo where the average rating of opponents is not used and provided every player has the same K factor it is always zero sum. FIDE rated events with players rated above and below 2400 are not zero sum and in fact there is no requirement for rating calculations to be zero sum. Glicko and Glicko2 are not zero sum.
    The Force can have a strong influence on the weak-minded.

  12. #87
    Illuminati Bill Gletsos's Avatar
    Join Date
    Jan 2004
    Location
    Sydney
    Posts
    16,544
    Seeing as it is totally misleading I'll remove the estimated rating change from the display.
    The Force can have a strong influence on the weak-minded.

  13. #88
    Reader in Slood Dynamics Rincewind's Avatar
    Join Date
    Jan 2004
    Location
    The multiverse
    Posts
    21,567
    Quote Originally Posted by Bill Gletsos
    In correctly calculated Elo where the average rating of opponents is not used and provided every player has the same K factor it is always zero sum. FIDE rated events with players rated above and below 2400 are not zero sum and in fact there is no requirement for rating calculations to be zero sum. Glicko and Glicko2 are not zero sum.
    Does it use the average rating to calculate the expected?

    I looked at the totals and the number of games between rated players was 311 but the total expected column was 316.71. Therefore, the calculation was expecting the pool to do 5.71 games more than is possible. This would lead to a rating deflation of the order of 57.1. The fact is was 59 was due to subsequent rounding error.

    The inflation in the expected score was (I thought) due to rounding the in the game by game calculation of the expected. However, if it is due to an average rating being used and multiplied by number of games played then that would also make sense as a number of players would have played one of the bottom 12 or so players which would drag their average down compared to the majority of the field they would have played. E.G. discounting my round 3 win actually increases my performance rating (over 10 games) as that opponent was nearly 400 points lower than any of my other opponents.
    So einfach wie möglich, aber nicht einfacher - Albert Einstein

  14. #89
    Illuminati Bill Gletsos's Avatar
    Join Date
    Jan 2004
    Location
    Sydney
    Posts
    16,544
    Quote Originally Posted by Rincewind
    Does it use the average rating to calculate the expected?

    I looked at the totals and the number of games between rated players was 311 but the total expected column was 316.71. Therefore, the calculation was expecting the pool to do 5.71 games more than is possible. This would lead to a rating deflation of the order of 57.1. The fact is was 59 was due to subsequent rounding error.

    The inflation in the expected score was (I thought) due to rounding the in the game by game calculation of the expected. However, if it is due to an average rating being used and multiplied by number of games played then that would also make sense as a number of players would have played one of the bottom 12 or so players which would drag their average down compared to the majority of the field they would have played. E.G. discounting my round 3 win actually increases my performance rating (over 10 games) as that opponent was nearly 400 points lower than any of my other opponents.
    When Swiss Perfect was written the FIDE code for rating gain was new rating = old rating + K(actual - expected) and the expected score was calculated by first determining the players rating - average rating of opponents, looking up this differnce in the FIDE rating table and converting it to a percentage score then multiplying this by the number of games played.

    I just now picked Peter Fry by random.

    16. Fry, Peter VIC 3074911 1908 5.5/10 7.10 1753 1789

    His rating is 1908 and the average rating of his opponents is 1753 for 10 games. That is a 155 difference.
    That equates to a percentage of 71%. He played 10 games so that gives an expected score of 7.1.

    Although Swiss Perfect defaults to a 350 point correction for the FIDE rating calculations it does not for the loacl rating calculations. I had changed this for the other lists I posted but missed it on this one.

    Therefore the average rating of Fry's opponents is only 1753 if no 350 correction is applied. He actually played a player rated 1360 and another rated 1555. These are both adjusted to 1558 once the 350 correction is applied and his average rating is thus 1773.

    As such the rating difference is 1908-1773 which equals 135 and equates to 68%. Hence his expected score is now 6.8.

    16. Fry, Peter VIC 3074911 1908 5.5/10 6.80 1773 1809

    However on a game by game basis with the 350 cutoff in place his expected score is 6.63 whereas his true expected score which is based on game by game with no cutoff in effect is 6.71.

    What this clearly demonstrates is how misleading calculations based on using an average can be whether they involve cut-offs or not.

    Note also that the 350 cutoff also affects his performance rating.

    His true performance rating is 1797.

    Note I have now corrected the above list.
    Last edited by Bill Gletsos; 14-01-2006 at 06:52 PM.
    The Force can have a strong influence on the weak-minded.

  15. #90
    Reader in Slood Dynamics Rincewind's Avatar
    Join Date
    Jan 2004
    Location
    The multiverse
    Posts
    21,567
    Quote Originally Posted by Bill Gletsos
    What this clearly demonstrates is how misleading calculations based on using an average can be whether they involve cut-offs or not.
    I agree. Means are meaningless.
    So einfach wie möglich, aber nicht einfacher - Albert Einstein

Thread Information

Users Browsing this Thread

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

Similar Threads

  1. Australian Schools Championships
    By jeffrei in forum Australian Chess
    Replies: 64
    Last Post: 27-02-2010, 08:57 PM
  2. paid CEO for Australian chess
    By chesslover in forum Australian Chess
    Replies: 15
    Last Post: 21-02-2006, 06:53 PM
  3. Mt Buller Australian Minor Chess Championships
    By cincinnatus in forum Mt Buller Chess
    Replies: 202
    Last Post: 24-01-2005, 02:44 PM
  4. Australian Clubs Team Champions
    By chesslover in forum Australian Chess
    Replies: 85
    Last Post: 26-02-2004, 06:59 AM
  5. Australian chess wheel reinvented again!
    By Kevin Bonham in forum Australian Chess
    Replies: 13
    Last Post: 20-02-2004, 04:01 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
  •