Yorkshire Chess Association

 I<< HOME

Year Book 2018-19 Contents

Notices

 

 Message from the President

Officers 2018-19

Annual Fees

County Match Fees & Petrol Allowance

Junior Contacts

YCA League Match Venues

Secretaries of Competing Clubs

Match Correspondents ‑ Woodhouse Cup

Match Correspondents ‑ IM Brown

Match Correspondents ‑ Silver Rook

YCA League Fixtures 2018-2019

ECF Game Fee Changes &c

Joining the ECF

Standard-play Grading Trends 2002-18

Notes on the YCA Grading List

Results Graded July 2017 to June 2018

YCA Grading List

Yorkshire Junior Reports

Correspondence Chess Report

U-160 Captain’s Message

2017-18 League Tables & Match Results

County Match Result Summary

English County Finals 2018

Recent Winners of YCA Events

Constitution and Rules

YCA League Rules

Index to Rules

Individual Championship Rules

Contact Details Index

Event Calendar 2018-19

 

26/01/2019

Yorkshire Players in the January 2019 ECF Grading List

 

An extract of “Yorkshire” players in the new January 2019 ECF Grading List in alphabetic order of name has been prepared and used the create three list:

List in Alphabetical Order of Name (Standard and Rapid)

List of Standard Rate Grades in Grade Order

List of Rapidplay Grades in Grade Order

 

Clubs have after them, in parentheses, a key to the Association/League in which that player represented that club in results on the ECF LMS or other source.

 

Of the 1103 “Yorkshire” names taken from the ECF grading list, 945 have January 2019 ECF grades of some sort, roughly equating to the 937 in the last YCA grading list.

 

Players in italics in the alphabetic list have yet to accumulate enough games in the ECF system to get a grade.  Some of these are players in leagues hitherto not ECF-graded, whose games prior to the current season are therefore not in the ECF system and so cannot be taken into account, but these players are likely to get a grade in 6 months' time, when more games have been accumulated.

 

The minimum number of games required is 5 in the same 36 months (of which at least 1 was in the last 12 months).  The number of games used is shown in the lists.  The number of games and period in which they were played is used to determine a “category” letter (A to F), which broadly imply decreasing degrees of inherent imprecision.  For details see ECF Grading website.

 

The grading calculation performed on a single game result yields a game-score which in itself is not any sort of measure of playing strength.  A “grade” is essentially produced by averaging out a reasonably large number of such game-scores.  Provided the set of opponents was fairly balanced as to stronger and weaker opponents, then that average is a reasonable measure of the players average playing strength over those games.

 

The traditional ideal minimum number of games perceived as yielding a reasonably accurate “grade” (give or take 3) is 30 games.  The fewer the games the less precise the “grade” can be purely on arithmetical grounds, quite apart from subtler statistical considerations.  However, the usefulness to congress organisers of approximate grades based on a single congress means grades based on as few as 5 games are published.  Note that adding a 6th game result to those five could alter the resulting “grade” by up to plus or minus 15, which serves as a measure of the degree of imprecision of grades based on so few games.

 

The more games used, the more precise is the calculation; yet the more games used, the less limited is the period of time to which the “grade” applies.  Conversely, using only recent games to give a “more up to date” grade is inherently more imprecise due to the smaller number of games.  Physicists might recognise here a parallel with Heisenberg's uncertainty principle.

 

Currently the ECF produces a grading list every 6 months, the July list (so as to be ready for the British Championships), and the January list.  Moving to monthly lists is being mooted, though the overall data-processing system might need to be upgraded in order to do that.  Monthly lists could be produced as at present, on the basis of a “rolling” 12-month data-collection period: July to June, then August to July, then September to August and so on.

 

A revised approach to grade calculation has been mooted, though whether the ECF has access to reliable advice on statistical matters, and whether they have access to adequate computing resources to set up and importantly maintain more-sophisticated systems are questions which arise.

 

Steve Mann