End of A 5 Year plan

[Phil Thomas]

FOLLOWING UP ANDREW'S SUGGESTION I FIND THAT I HAVE MADE A SERIOUS ERROR.

WHAT I THOUGHT WERE DECEMBER 2006 GRADES WERE IN FACT JULY 2006 GRADES.

SORRY EVERYONE

LOOKING DEEPER INTO THE DATA THIS CHANGES FINDINGS SUBSTANTIALLY AND RAISES SOME INTERESTING QUESTIONS.

RATHER THAN CLOG UP THIS POSTING I WILL CREATE A NEW ONE TO EXPLORE THE CORRECTED DATA.
THIS WILL APPEAR LATER TODAY

WHEN TIME (DOUGIE AND MINE) ALLOWS INSTEAD OF PUBLISHING NUMBERS 5 YEARS APART AND ASSUMING LINEAR TEND BETWEEN THE POINTS I WILL PUBLISH DATA AS OF JULY 2006, 2007, 2008, 2009, 2010, 2011

End of Correction


SECOND EDIT DEC 20th 2011
WITH THANKS TO DOUGIE BRYSON I NOW HAVE GRADING DATA FOR THE START OF SIX CONSECUTIVE SEASONS JULY 2006 TO JULY 2011. THESE I HAVE SUMMARISED IN A NEW POSTING AT THE END OF THIS THREAD.

I LOOK FORWARD TO COMMENTS FROM ALL THE USUAL SUSPECTS AND PERHAPS FROM SOME UNUSUAL SUSPECTS.

**********************************************************************************

5 years ago in December 2006(EDIT of Dec 13th DEC 2007) my son Daniel was in Primary 7. He was the only primary school pupil in Scotland with a grade above 1000. There were at that time very few S1, S2 and 3 pupils with a grade.
Using the excellent on line grading system I recorded the 10th highest grades in each J category.

Exactly 5 years later it is interesting to note the differences made as a result of (a) the increase in Junior competitions notably SJC and Hamilton events and (b) the initiatives I put in place and actions I took as IJD between 2008 and 2011.

Here then are the two sets of figures and for convenience the increases made over this 5 year time frame.

Dec 2006 tenth highest grade
J7 less than 10 graded players
J8(and below): 183
J9: 440
J10: 578
J11: 798
J12: 972
J13: 1005
J14: 1261
J15: 1485

Dec 2011 tenth highest grade

J7: 183
J8(and below): 355
J9: 527
J10: 663
J11: 886
J12: 1100
J13: 1186
J14: 1368
J15: 1530

INCREASES OVER 5 YEARS
J8(and below): 172
J9: 87
J10: 85
J11: 88
J12: 128
J13: 181
J14: 107
J15: 45

ALL POSITIVE NUMBERS IN NONE OF THESE AGE GROUPS HAS THE 10th HIGHEST GRADE DECLINED

[Andrew McHarg]

Hi Phil

Out of interest; why did you use the tenth highest grade? I can see why you wouldn't use the top grade for statistical reliability but surely an average would give a more reliable figure? Or am I missing something?

[Angus McDonald]

Phil,

Out of interest could you tell me what Ian's grade was at the same date you started your five year plan?

TIA,

Angus

[Phil Thomas]

Hi Andrew,

Fully agree with you. Averaging the top 10 grades would have similar or enhanced validity to looking at grade number 10. Either method could have been used. Pragmatically it was quicker in 2006 to extract a single number per age group.

The choice of 10th was not arbitrary. In theory bottom board at the Liverpool Quadrangular U12 and U14 teams approximates to two of those 10th places. In practice grade is not everything and confusingly Liverpool Quad uses Sept 1st for age cut off.

Presumably the grading system has archived data to allow comparisons 5 years apart - but this might have to be July 2006 vs July 2011. Dougie if you're reading could you send me excel file(s) to allow me to extract that data.

My intuition is that 5 Year grade increases would look more impressive due to the difference between 10th place and the top 2 or 3 places increasing over recent years.


POST SCRIPT,
Angus,

Just read your post.
5 mins to end of my lunch break so brief reply only
Would the grading page take you straight to that number? It will certainly show you a graph of Ian's grade changes over the last few years.

[Angus McDonald]

Phil,

Grading graph doesn't appear accurate to me!? ( perhaps a year out of sync?)

I think Ian i.e was about 950 grade at end of 2005/2006 season and about 1300 grade at end of 2006/2007 season. I don't recall him being graded lower than Daniel at any point during the 2006/2007 season. If he was it was a very short period of time perhaps Dec 2010?? That's why I was wondering if you'd logged Ian's grade at the time.

Angus

[Andrew McHarg]

I'm not sure that the graph is out-of-sync, but I can check it by the end of December. I intend to put a table of actual grades over the years on each person's stats anyway and from there it will be obvious whether the graphs are entirely accurate. The way the graphs work is a bit clunky, so it wouldn't overly surprise me if there were innacuracies. ;P

[Angus McDonald]

Andrew,

That sounds like more work for you!

It's important though that if you are going to hang your hat on statistics then they should be accurate or there is little point. imho

Allthough all the graphs really look nice Actually I think they are also accurate with only possible blip being the linear graph of grading performance.

Also though when you look at an individuals past games it doesn't appear to give their grade for that year and how it was calculated. On the other hand I could be reading it wrong.

[Mike Scott]

In season 2007 - 8 (which I think Phil is really meaning as my Jonny was P7 in 2006 Dec.) the J11 (and those normally in P7) top 3 players based on published grades:-

Thomas Daniel 1051
McDonald Calum Fraser 1099
McDonald Ian 1329

Interestingly the 10th J16 in 2007-8 had a published rating of 1237, in 2011 - 12 ? 1166 so that is a -71 pts drop.

You certainly put a lot of hard work and time as IJD - and we remain grateful for the chances this gave Jonny -, and as part of the SJC organising events, but I am not sure exactly what initiatives and actions as IJD you are referring to that would have contributed to the claimed improvements amongst the lower rated juniors who would never attend an international event.

[Andrew McHarg]

That sounds like more work for you!

It's important though that if you are going to hang your hat on statistics then they should be accurate or there is little point. imho

Allthough all the graphs really look nice Actually I think they are also accurate with only possible blip being the linear graph of grading performance.

Also though when you look at an individuals past games it doesn't appear to give their grade for that year and how it was calculated. On the other hand I could be reading it wrong.

It is a wee bit more work, but it's one of those things where the foundations are already in place. Finding space for it without it looking out of place will be harder than programming it.

Wasn't it Bill Gates who once said: "If you can't make it good then make it look good". The result: Microsoft Windows! :U In reality though; I agree with you. ;P

Others had mentioned about the grades not showing historical calculations. I'll get around to sorting it. It's not as easy as it initially appears because of some stupid technical reason but I'll eventually devise a way. 8)

PS: How did Iain get so good so quickly? I always remember him being good, but comeon! What's the secret? A book? A diet? Bad Aberdeen Weather? :\

[Phil Thomas]

In season 2007 - 8 (which I think Phil is really meaning as my Jonny was P7 in 2006 Dec.) the J11 (and those normally in P7) top 3 players based on published grades:-

Thomas Daniel 1051
McDonald Calum Fraser 1099
McDonald Ian 1329

Interestingly the 10th J16 in 2007-8 had a published rating of 1237, in 2011 - 12 ? 1166 so that is a -71 pts drop.

You certainly put a lot of hard work and time as IJD - and we remain grateful for the chances this gave Jonny -, and as part of the SJC organising events, but I am not sure exactly what initiatives and actions as IJD you are referring to that would have contributed to the claimed improvements amongst the lower rated juniors who would never attend an international event.

Hi Mike,

you have spotted my typo.

Daniel was in P7 in Dec 2007 (and P6 in Dec 2006)

Angus Ian is I recall 10 months older than Daniel with a January Birthday, hence went into S1 a year earlier than Daniel.

Just double checked my memory with Primary Individual Chess Championship results 2008

Daniel's highest graded opponents were.

Calum Birkett 989
Peter Rainey 978
Kiron Roy 914
Connor MacKay 907