Birth year of chess players

[Roger de Coverly]

It is crucially important that the graders obtain dates of birth for any player without a grading code, especially juniors.

The grading system can work quite happily without and has done so for many years. For new players, the more crucial information is whether this is their first event, so they don't get duplicated. For existing players, the critical information is their grading code. Even for juniors, you only need the year of birth and whether they were born before or after 1st September. Actually I'm assuming that "age" for the age related increment purposes is always the same on the January list as the July/August list.

[Adam Raoof]

It is crucially important that the graders obtain dates of birth for any player without a grading code, especially juniors.

The grading system can work quite happily without and has done so for many years. For new players, the more crucial information is whether this is their first event, so they don't get duplicated. For existing players, the critical information is their grading code. Even for juniors, you only need the year of birth and whether they were born before or after 1st September. Actually I'm assuming that "age" for the age related increment purposes is always the same on the January list as the July/August list.

"PLAYER IDENTIFICATION

A major problem experienced by graders is the identification of players. Players with similar names are sometimes wrongly conflated. More often, a single player acquires two (or even three) identities because no one is sure whether it is the same person each time. It is helpful if:

1) Event organisers and match captains ensure that they report: players’ forenames (not just initial and surname); their date of birth if known; and, where possible, their Grading Code from this List. In the case of new players, congress organisers must report the club (or place of residence). For existing players, if they report the club they must ensure that it comes from the player himself. Not from this list. Clubs in this list may be out of date.
2) Players make sure that event organisers have their correct details. If moving to a different area, or playing in a new event, players should ensure that match captains and congress organisers know where they are from and enter their Grading Code and full name on result sheets and pairing cards (or in their computer files as the case may be).
Date of Birth and Age

Date of birth is very helpful in player identification. Additionally, as the junior grading enhancement is based on age, it is especially important that we have the DoB for juniors. (See below: How ECF Grades are Calculated)

Date of birth is not published. Age is published for Juniors only. For a July or August list it is their age on 31st August in the year of publication. For a January list it is their age on 31st August preceding publication."

[E Michael White]

It is sometimes said the DOB is needed to differentiate between players with the same name and initials. Well firstly it wont normally help if they are twins Alan and Adrian and secondly a better method is to ask for an alternative secondary secret which does not have security implications.

[John Foley]

That's a nice graph, but I'm trying to relate it to my own grading history. In 2002 I was ECF 145 and aged 24/25. In 2012, I'm 34/35 and ECF 171 (allegedly). Is it not possible to do further analysis and determine whether the 'peak' age is different for those with lower and higher grades?

The graph shows the average grade for each age range. It is a cohort analysis. What you are looking for is to be able to aggregate individual trajectories. This would require the grade history by individual. The ECF grading database is only populated since 1994 so you could only analyse those people who have been graded for 18 years i.e. those who are aged below around 30. By coincidence, this is the age range in which we might expect to see the fastest rise in gradings. It would be interesting to view the dynamic first stage of the individual chess life cycle. The ECF database will continue to accumulate information and in about 40 years time you would be able to obtain a complete answer to your query. In the meantime, the only way to answer your question would be to construct a sample. Personal grade histories over complete lifetimes obtained from historic sources would be required.

There are several hypotheses to explain the data pattern in the infographic and further analysis would be required to test them. One approach would involve quantifying the following categories:

a) DOWN SHIFTERS

This refers to those people who have the leisure time to return to chess, or take it up as a hobby. Retirement/redundancy/career change may trigger this shift. Chess returners may have lower gradings than when they were younger. Hobbyists may never develop a higher grading. This category is numerous and brings down the average grading.

b) TALENT SHIFTERS

This refers to those people who are very talented at the game and have a higher than average grading but choose to leave chess to pursue other interests in life. Being talented at chess is often correlated with being talented in other walks of life. As they leave the database, the average grading declines.

c) STEADY IMPROVERS

These are the people who never had a chess fallow period but gradually improved their performance level over the years through experience, practice and study. This relatively small category comprises those few who have become relatively strong players as well some dedicated chess lovers. Their numbers cannot be significant or it would be reflected in the graph.

Can anybody suggest other categories?

[Roger de Coverly]

Additionally, as the junior grading enhancement is based on age, it is especially important that we have the DoB for juniors.

For the use the ECF makes of it, it only needs the year of birth and whether the birth date is before or after 1st September. So going back forty years, I would have attained the age of 12 in 1962. So on the 1st September 1962, I would either have become 12, in which case that's my BCF age for 1962-63, or I would be yet to reach my birthday, so my BCF age would be 11. Day and month of birth aren't needed to establish my BCF age.

[Alex Holowczak]

More often, a single player acquires two (or even three) identities because no one is sure whether it is the same person each time.

That's if you're lucky! A localish junior, with a fairly common surname, started playing in events all over England. The then Manager for Grading was faced with merging not two, or three, but seven instances of the same player.

I've had to inform Richard of a few merges that need to be made for precisely this reason. There was also a player in the list as Player, Anthony and Player, A; one from a league where you only report initials and not forenames, and so two separate people were created. However, this would also have been avoided if the grader of the initials league had updated his masterlist.

[Paul Cooksey]

I'm inclined to tend with the keep personal data to a minimum arguments.

But this is a historic argument isn't it? We will have universal (/compulsory) membership soon, and people will be quoting their membership number whenever they enter a competition won't they?

(I'm not sure we even need separate grading and membership numbers any more)

[Roger de Coverly]

c) STEADY IMPROVERS

These are the people who never had a chess fallow period but gradually improved their performance level over the years through experience, practice and study. This relatively small category comprises those few who have become relatively strong players as well some dedicated chess lovers. Their numbers cannot be significant or it would be reflected in the graph.

Have you noticed that the BCF/ECF allocated grading codes in sequential order and has done since around 1986? There's quite a number of players still playing with old "eighties" code numbers. That's not to say that all of them have played continuously since then. Believers in grade deflation would have us believe that their grades fell over this period. At the more elite end, that's really not true and players 160 plus in that era who have played consistently, keeping up to date, are still in that range.

[Roger de Coverly]

But this is a historic argument isn't it? We will have universal (/compulsory) membership soon, and people will be quoting their membership number whenever they enter a competition won't they?

It isn't proposed at the moment that membership will be compulsory for tournaments. Imagine the noise from north (or even west) of the Border if Scarborough or Blackpool demanded ECF membership as a condition of entry.

[Paul Cooksey]

It isn't proposed at the moment that membership will be compulsory for tournaments. Imagine the noise from north (or even west) of the Border if Scarborough or Blackpool demanded ECF membership as a condition of entry.

Does the ECF need to care about the ages of players from other federations? A competition might, to determine eligibility, but I don't think they would need to pass PID on.

[Roger de Coverly]

Does the ECF need to care about the ages of players from other federations? A competition might, to determine eligibility, but I don't think they would need to pass PID on.

The calculation and data methodology of the ECF domestic grading system is blind to the Federations of the players taking part. So it needs date of birth for a new (to the ECF) Scottish or Welsh player just as much or as little as for a new English player. That's not the case in Wales and Scotland where players new to their systems would have a different treatment if they already had an international rating or ECF grade.

[Christopher Kreuzer]

That's a nice graph, but I'm trying to relate it to my own grading history. In 2002 I was ECF 145 and aged 24/25. In 2012, I'm 34/35 and ECF 171 (allegedly). Is it not possible to do further analysis and determine whether the 'peak' age is different for those with lower and higher grades?

The graph shows the average grade for each age range. It is a cohort analysis. What you are looking for is to be able to aggregate individual trajectories. This would require the grade history by individual. The ECF grading database is only populated since 1994 so you could only analyse those people who have been graded for 18 years i.e. those who are aged below around 30. By coincidence, this is the age range in which we might expect to see the fastest rise in gradings. It would be interesting to view the dynamic first stage of the individual chess life cycle. The ECF database will continue to accumulate information and in about 40 years time you would be able to obtain a complete answer to your query. In the meantime, the only way to answer your question would be to construct a sample. Personal grade histories over complete lifetimes obtained from historic sources would be required.

Thanks for that explanation. Is the data from before 1994 available anywhere? I had thought that was the year of my first grade, but maybe not.

There are several hypotheses to explain the data pattern in the infographic and further analysis would be required to test them. One approach would involve quantifying the following categories:

a) DOWN SHIFTERS

This refers to those people who have the leisure time to return to chess, or take it up as a hobby. Retirement/redundancy/career change may trigger this shift. Chess returners may have lower gradings than when they were younger. Hobbyists may never develop a higher grading. This category is numerous and brings down the average grading.

b) TALENT SHIFTERS

This refers to those people who are very talented at the game and have a higher than average grading but choose to leave chess to pursue other interests in life. Being talented at chess is often correlated with being talented in other walks of life. As they leave the database, the average grading declines.

c) STEADY IMPROVERS

These are the people who never had a chess fallow period but gradually improved their performance level over the years through experience, practice and study. This relatively small category comprises those few who have become relatively strong players as well some dedicated chess lovers. Their numbers cannot be significant or it would be reflected in the graph.

Can anybody suggest other categories?

I would also look for those that enter over-graded and then decline to their actual level, and those who enter under-graded and ascend to their actual level. Or rather, check that this adjustment doesn't take several years (I suspect it does). You also get people entering from other grading systems. I played a player last night from Scotland, studying in London, and he has an ECF grade on the July 2011 list and the January 2012 list (both grades are 200+ with some pretty impressive results). I was surprised there was no more grading history than that, but then realised the earlier history is probably in the Scottish grading system. I just looked and found someone with the same surname of about the right age, but can't access the details. Same player also appears in the FIDE rating list. What I'm trying to work out is whether the three grades (FIDE, Scottish, ECF) are very different, and why that might be (what I can't seem to access is the Scottish grade).

You might also get those that enter at the right level and never change much over the years. Some of the elite slot in at the top and stay there, for example. You also have the adjustment that was made and possible grade inflation. I suspect my personal grading history is a combination of improvement in the early years, gaining experience as that improvement boost ran out, plus some inflation and the adjustment.

A more detailed analysis would look at whether players were playing those at the same level as them, lower, or higher. Someone who consistently beats those lower graded might go up in grade but be unable to cope with playing those at that new level. Conversely, someone playing those graded much higher might lose lots of games, go down in grade, and then start playing those lower graded and go back up again. That's different to playing those of around the same level, getting a slight plus or minus score, and slowing increasing or decreasing in grade. When that starts to happen, you've probably reached your level. But you won't know that unless you try playing up or down levels.

[Roger de Coverly]

(what I can't seem to access is the Scottish grade).

The Scottish ratings are published, but you won't be able to see records detailed by game without being a Chess Scotland member. It's quite obscure how to use their grading website. It's come up before, here's a thread that might help
http://www.ecforum.org.uk/viewtopic.php?f=4&t=3415

[Christopher Kreuzer]

(what I can't seem to access is the Scottish grade).

The Scottish ratings are published, but you won't be able to see records detailed by game without being a Chess Scotland member. It's quite obscure how to use their grading website. It's come up before, here's a thread that might help
http://www.ecforum.org.uk/viewtopic.php?f=4&t=3415

Thank-you, I'd forgotten about that.

So, I now have the Scottish grade (1931) but not the history. The ECF grade (209 in July 2011 and 213 in January 2012, with the history), and the FIDE rating (2048). I'm now trying to find various conversion formulae...

Found one here:

http://www.chessscotland.com/forum/view ... f86bb4e959

So which conversion formulae should I be using? I'm confused. I suspect the ECF grading history of this player is probably the most current and accurate, but if I apply some of these conversions, I get:

ECF 213 *10 +180 = CS 2310
ECF 213 *8 +650 = FIDE 2354
(CS 1931 -180)/10 = ECF 175
(FIDE 2048 -650)/8 = ECF 175

Which actually does help somewhat. Scottish junior comes south (some pretext about studying at university) and causes carnage in the London leagues, hoovering up grading points right, left and centre. Which category does that fit in, John?