Should junior only events have a separate grading system?

[Neill Cooper]

Discuss!

[Mick Norris]

Yes

[Matthew Turner]

There would be potentially be several advantages to a junior grading list.

1. You could have an ELO system so beginners would start with a grade of 400 or whatever which sounds better than 7
2. It could be updated regularly
3. It could be more flexible than the main system, so allow for players to be entered onto the system with an estimated grade

We have seen in another thread the attitude of ECF officials to junior chess and really this is an opportunity to take junior chess forward by establishing a junior organisation for the benefit of juniors. How grading junior games is paid for is something of a secondary issue, game fee would work, but membership would be better. No compulsory membership required, just if you pay you get a grade and perhaps other benefits, if you don't you don't.

[Alex McFarlane]

I have long held the view that there has to be progression in chess.
As far as grades are concerned historically I would see this as Rapidplay Grade as the first target. After a bit of improvement/consolidation the next target would be a national grade. This would be followed by a FIDE rating. Unfortunately, FIDE have spoiled this by lowering the floor to 1000.

My concerns about attaching numbers to juniors (and not just juniors) is that they develop a mind-set of protecting their grade rather than improving. Improving in chess means experimentation with different openings and playing in an occasional stronger event (not simultaneously). Both of these will mean that the grade will stagnate or decline. Obviously this can be a problem. I always spoke to parents of kids with potential to warn them of this in the hope that they could prepare their child for this.

Children love to have a measure of success. A grade provides this measure. But care should be taken – it can also be a measure of failure. If you do have a junior list I would suggest a maximum published figure. Children nearing this should be encouraged to play in 'adult' events.

When I ran a school club the children certainly liked the fact that I would calculate their grade after every club meeting and give bonus points for their performance in school matches. I used a three figure grade as this meant that when they progressed to external chess tournaments and the four figure Scottish system even a low figure could be seen as an improvement on the school figure. (Whilst the grading list appeared only once a year the Chess Scotland website provides a performance rating for every event.) The Scottish Junior Chess Association also published a bi-monthly list for a number of years (again using a three figure system if memory serves). This was a big incentive to the youngsters at the time but was overtaken by the performance ratings given on the CS website. When the person who collated the list resigned he was not replaced.

A word of caution which applies to any grading system. Where there is a number of effectively independent areas with little cross play between them, the figures will quickly become skewed. A reasonable player in one relatively weak area can easily get a high grade and appear to be a better player than a stronger player from an area which is more competitive. I can remember many years ago considering selections for a Scottish Primary team. The Isle of Lewis would have provided a significant number of team members if selected by grade. Because of their isolation the players were unknown to the selectors. 4 or 5 of them were invited to the mainland for a tournament. None of them performed to anything like their grade and only one showed any real potential.

[Roger de Coverly]

No compulsory membership required, just if you pay you get a grade and perhaps other benefits, if you don't you don't.

That approach requires an Elo style system. So you need a rule which says that games only count for rating if they are played against players already with a rating, which ensures non-members never get a rating. You still need a rule to handle games against ex-members or visiting players not required to be members, whether they should be ignored or calculated on a one-legged basis.

In the context of a Junior rating system, you might give new players, assuming they qualified for one, an arbitrary age based initial rating.

[Martyn Harris]

Yes because a) including games from junior only events may undermine the integrity of the main grading system, and b) grading systems (ECF or ELO) are not designed in a manner which offers encouragement to very weak or slow to improve players. There may be a case for abandoning any idea of a grade purporting to show standard here, and instead looking for an upward moving only number based on cumulative achievement.

[Roger de Coverly]

There may be a case for abandoning any idea of a grade purporting to show standard here

Objectively grades only ever supply a means of calculating ranking. They only show standard if you can keep them stable over time.

The other viewpoint of Junior Organisers might be that there are numerous players and they have no idea who the best ones are. Whilst an individual tournament will give you a ranking for that tournament, you need several and a means of linking them if the participants aren't the same. Grading or rating are the means of linkage, but you need a system whose rules rate by event rather than individual player.

[Martyn Harris]

Perhaps one needs to ask whether the grades are for the benefit of the players or the organisers.

[Andrew Zigmond]

Perhaps one needs to ask whether the grades are for the benefit of the players or the organisers.

A bit of both.

I remember getting my first published grade when I was about 15, which was a grade of 46 based on several years of playing my clubs minor tournament where obviously I lost a lot of games to more experienced players. I remember being quite angry at the time that this grade didn't reflect a) my true strength (unless you played thirty games in a season the grade was diluted by games from past seasons and six games per year wasn't a lot) and b) the improvement I felt I was showing. It did however give me a mark against which to improve. At a school level a rating list would allow organisers to introduce players to games played under match conditions rather than endless knockabout friendlies.

Once a junior progressed to the adult list they would need to be warned that they would be starting at the bottom again and improvement might not be reflected immediately.

The question is who would organise such a list. Either the ECF in which case they would expect a fee to be paid, or an independent list that may challenge the authority of the ECF.

[Krishna Shiatis]

No.

You need a unified system with the adults to avoid confusion, measure progression and also to keep track of everyone (more useful to the people at the top I know, but useful nonetheless). Also, what would happen if kids played in both adult and junior - how would their grade be measured - would they hold two separate grades?

It is not just for the juniors to understand but also their parents. Not to say that they could not follow two separate systems, but it would mean all sorts of issues when organising board order and pairings if some kids use one grade and others another.

In addition, if some children have played in junior only events and then turn up to adult events, you would need to have some form of conversion. These kind of things can cause controversy when prize money comes into the equation if people think they have been hustled in some way.

That is not to say that a unified system does not have its own issues, but it makes more sense to improve what we have already rather than separate juniors from adults.

[Roger de Coverly]

Also, what would happen if kids played in both adult and junior - how would their grade be measured - would they hold two separate grades?

That would be the point, that the adult grade represented only play in adult events, so high Junior grades earned solely by using the 40 point rule against weak opposition in Junior tournaments would no longer appear alongside adult grades.

Particularly if Junior events were run by a separate national organisation, it would take them out of membership rules framed on the apparent assumption that everyone belongs to a club playing league chess with twenty or more games every season and no incentives are needed to attract new players.

[Paolo Casaschi]

No.

A rating/grading system is more accurate the larger the pool of games considered. Fragmenting into several separate grading/rating systems would make each of those less accurate than what you would have with a single unified system.

I agree with Alex McFalrlane concern ("My concerns about attaching numbers to juniors (and not just juniors) is that they develop a mind-set of protecting their grade rather than improving."); it probably applies to adults as well, including myself. However, I believe the solution should have more to do with educating the players (both junior and adults) about the value of the rating/grading number rather than artificially tweaking the number itself.

[Brian Valentine]

Whether a separate grading system is good for juniors or not, one needs to consider whether this is a good idea for the system.

The ECF grading approach relies on a steady replacement of leavers. If a significant proportion of junior games were excluded this is likely to be inflationary as improvers will be sucking less points out of the system.

All common rating systems are based on the premise that they are trying to estimate a relatively stable underlying skill. A junior system should be based on an underlying assumption of improvement and this could lead to a different approach.

However I am in the camp that thinks two systems in England is one too many (even if either of the alternatives could be improved especially with regard to the treatment of juniors).

[Roger de Coverly]

The ECF grading approach relies on a steady replacement of leavers. If a significant proportion of junior games were excluded this is likely to be inflationary as improvers will be sucking less points out of the system.

If the intention is to remove only games and players in exclusively junior events, how much impact does this really have? It removes the problem of zero or negative grades. With the feature of treating juniors as new players every season, are they still sucking points out of the system? Arguably the flat 10 points for every under 18 that used to be used in the 1970s and 1980s injected points. Or at least that was what was supposed in the late 1980s when it became established that the mean and median grades were increasing.

[Neill Cooper]

The majority of graded junior only events are rapidplays.
Of the 93215 radpidplay half games last year half (45630) were in junior only events.
Of the 4685 players with published radidplay grades, half (2487) were juniors.
Data thanks to Richard Haddrell's excellent analysis at http://www.sccu.ndo.co.uk/grad.htm