19-06-2013, 12:17 PM
Patrick McGovern Wrote:Given the major changes that are mooted for the council meeting would it not be better to give them due consideration at the AGM? We could have the situation when policy is changed (or not) and decision then reversed at AGM.
An excellent well thought out post from Pat. The motion on junior selection runs to a 10 page pdf. File. There is inadequate time to discuss on the notice board before the council meeting and one doubts that there will be time during the council meeting.
Totally agree with Sean about the need to test the system by trying it out his methodology on juniors that have played at least 4 events. I was surprised though that only a single player was selected to test the system (Murad).
The motion tells us that Murad had 4 congresses at which he graded out at 1924, 1812, 1829, 1910. Mean of these 1869. Very close to live grade of 1848. Hence the grading system works.
However, looking up Murads Scottish events today I am puzzled to find on the grading page 4 different numbers 1663 (Ayr) 2065 (Edinburgh) 1654 Hamilton 2180 (Dundee). The mean of those 4 is 1891. Which again is close to the live grade of 1848.
All of which merely tells us that the grading system is working.
Let us do a mind experiment. Suppose Murad played another 4 congresses and scored identical results to the first 4. His strength would then be evaluated from these 8 numbers.
1663, 2065, 1654, 2180, 1663, 2065, 1654, 2180.
The correct way to do this would be to average all 8 – result is still 1891.
The incorrect way to do this is to average the highest 4 and get 2123.
Where did the extra 232 points come from ?
Its not magic its a serious mathematical flaw.
Cherry picking data this way gives, in effect, a set of corrupt raw data.
I wondered why the grade of player number 1 on the list was not subjected to the same treatment. From the grading database Andrew McClement has 2 tournaments -the Scottish and the 4NCL in England.
Tournament performances were 2129 and 2368.
The mean of those 2 numbers is 2244 - significantly higher than live grade.
It concerns me that the example given for Monica Espinosa, suffers from the input of poor quality raw data.
For Monica on the grading database I can see, today ,4 tournaments with grading results of 1663(Ayr) 1347 (Edinburgh) 1621 (Glenrothes) 1375 (Girls Championship).
The data actually entered was 1454 1428 951 with no fourth tournament.
Can somebody please tell the notice board readers where that number of 951 came from? If it was a typo it needs to be corrected.