x
Chess - Play & Learn

Chess.com

FREE - In Google Play

FREE - in Win Phone Store

VIEW

How To Rate Kid Players?

  • #1

    I vounteer at an elementary school and we have about 140 kids enrolled in our chess club. We would like to be able to rank them somehow, so next year they will be divided into categories like Beginner, Intermediate and Advanced. (There may be other subdivisons, if it makes sense.)

    Of course, if you ask a six year old what level they are, they typically answer "advanced," even if they can't set up the board.

    We are looking for a quick and easy system to help us classify the players. Of course, we have little or no money for this task, and our time with the kids is limited to about half an hour per week.

    Is there a website or software that does basic chess rating for beginners? Or is there simple criteria that we can use to categorize the players?

    Any suggestions?

  • #2

    Normally if they are provisional at least is canada they start at 1500 and they have to play 25 games to get an actually rating.

  • #3

    USCF rating system

  • #4
    2200ismygoal wrote:

    Normally if they are provisional at least is canada they start at 1500 and they have to play 25 games to get an actually rating.


    yes that is right i am a kid and play in tournaments. You start with 1200

    there is a free site that let you rate them here it is :9it is in french,sorry!) http://www.fqechecs.qc.ca/index.php?typ=acteur&sc=fqe/bdcotes&r=cc

  • #5

    We are looking for more of a "rough cut" so we have an idea of where kids stand within our group.

    So another way to look at the problem is: Given 140 kids, how could you quickly divide them into groups so they are playing opponents that are about their level.

    We can't really go by age, because some of the first graders are better than some fifth graders. Some kids have been playing for years, but not consistantly.

  • #6
    sata123a wrote:

    USCF rating system


    One thing to be verycareful of, if you're going to implement an independent  ELO or Glicko style rating for a small pool like this, is that it will in effect be an isolated pool and the ratings you arrive at will apply only to that pool and can't be reliably compared to other ratings.

    If you've got one kid who consistently beats every other kid in the program, he may well end up with a rating of 2800 -- this is totally appropriate for the pool, but by no means makes him a grandmaster.  This is an expectation you'll have to manage carefully by explaining why a rating of 2800 in an isolated pool of 25 elementary school students is not the same as 2800 FIDE.

    Incidentally, this is a misconception that also plagues chess.com.

  • #7
    Bradison wrote:

    We are looking for more of a "rough cut" so we have an idea of where kids stand within our group.

    So another way to look at the problem is: Given 140 kids, how could you quickly divide them into groups so they are playing opponents that are about their level.

    We can't really go by age, because some of the first graders are better than some fifth graders. Some kids have been playing for years, but not consistantly.


    How about setting up a perpetually running ladder tournament? 

    http://en.wikipedia.org/wiki/Ladder_tournament

  • #8
    p { margin-bottom: 0.08in; }

    A ladder might work, but we would probably implement it next year. One problem with a ladder approach, is that kids can elect to come to chess club, or play outside. So given any week, we don't know how many kids will show up. It's a fairly loose crowd.

    Last year we had a bead reward system. If you won, you got a bead. But kids seemed to be more focused on the beads than chess, and it lead to some unwanted behavior around the beads themselves.

    We had considered asking kids if they know how to set up the board, or castle, or understand other advanced moves. But I think we need some kind of quick litmus test, rather than asking a yes or no question. The other issue with this approach is that kids could jump a level just by learning a new chess move.

    We don't have an ideal situation, so I don't know if there is a "nice" solution, but all of your ideas are helpful.

  • #9

    #1 : How about a  "random within" Swiss pairing tournament

    A over-simplified (winging it) way of doing this would be to take advantage of the swiss system ...

    1) Schedule a tournament of say 5-8 rounds of quick 5-min games.

    2) Pair as randomly as you like for round 1

    3) After round group, group the winners and losers into two groups (draws could be randomly split)

    4) Repeat random pairings from within the sub-groups

    5) Now you'll have 5 sub-groups, those with 2 wins, 1.5 wins, 1 win, 0.5 wins, 0 wins etc.  Keep re-grouping and pairing within after each round.

    6) Assuming 5 rounds, you'll have the diff. "classes of players" at the end of the day

    The rationale for this approach is that if the pairing is random, better chess players tend to "stand out" by winning consistently more than they lose. By making winners play winners, you are further "filtering" the pack to see who dominates.

     If you want to factor out bad luck (the top 4 players of the entire group all play each other too soon), increase the number of rounds  and you should be able to place the kids in fairly graded containers of chess skill.

    For instance:

     

    Scored / Categories of Skill?

    4-5 : Class A (for lack of a better group name)

    2-3.5: Class B

    0.5 - 1.5 : Class C 

    0 : Class D 

    #2 : IF this is too elaborate to implement, then perhaps a written quiz with a bell-curved difficulty spread across the questions? 

  • #10
    Shivsky wrote:

    #1 : How about a  "random within" Swiss pairing tournament

    A over-simplified (winging it) way of doing this would be to take advantage of the swiss system ...

    1) Schedule a tournament of say 5-8 rounds of quick 5-min games.

    2) Pair as randomly as you like for round 1

    3) After round group, group the winners and losers into two groups (draws could be randomly split)

    4) Repeat random pairings from within the sub-groups

    5) Now you'll have 5 sub-groups, those with 2 wins, 1.5 wins, 1 win, 0.5 wins, 0 wins etc.  Keep re-grouping and pairing within after each round.

    6) Assuming 5 rounds, you'll have the following "classes of players" and the end of the day

    The rationale for this approach is that if the pairing is random, better chess players tend to "stand out" by winning consistently more than they lose. By making winners play winners, you are further "filtering" the pack to see who dominates.

     If you want to factor out bad luck (the top 4 players of the entire group all play each other too soon), increase the number of rounds  and you should be able to place the kids in fairly graded containers of chess skill.

    For instance:

     

    Scored / Categories of Skill?

    4-5 : Class A (for lack of a better group name)

    2-3.5: Class B

    0.5 - 1.5 : Class C 

    0 : Class D 

    #2 : IF this is too elaborate to implement, then perhaps a written quiz with a bell-curved difficulty spread across the questions? 


    +1

  • #11

    Shivsky, you have some great ideas. I'm going to run them by our group and see if we can make them work.

    We would need to work out the logistics of a tournament ranking given our time constraints, but it sounds like a sound suggestion. I think if we make the decision to rank via tournament play, then the "random within" idea will be a strong contender.

    We had never considered giving a written test, but now that you've planted the idea, it opens a world of possibilities.

    Thank you.

  • #12

    Tournament is a good idea.

    Otherwise hold a simul - (might need more than one person dong the rounds for 140).

    Those that last longest probably better than those that don't. Else the Adult could fill in some assessment criteria during and after the game for each kid - and rate them according to the assessment.

    But the tournament the better idea.

  • #13

    The only real drawback that I can see with doing a tournament is that we will probably have a bunch of kids who don't show up on that particular day. When they show up the following week, we would have to somehow have to rank them into the group that has already been ranked.

  • #14

    How about getting 5 positions and asking each one to solve them?

    A very easy mate in 1, then a simple fork, then a mate in 2, then a discovered attack winning a piece, then a mate in 3 or something more difficult.

     

    Should be easy to separe the kids who can barely move the pieces right from those who are more experienced.

  • #15

    Why would USCF Classes have anything to do with that specific pool of players?

    Also, he mentioned that he could not find a more approapiate name for the group, making it even clearer that the naming has nothing to do with the FIDE or USCF classes.

  • #16
    TheMouse wrote: Didn't Claude Bloodgood exploit this?

    Yes, this is exactly the exploit that he used. 

  • #17

    Before I clicked this thread, I genuinely thought you wanted to rate players who used the King's Indian Defence. Ah well...

    I think a ladder or swiss tournament (both of which have already been suggested) are the ideal solutions to the problem.

  • #18
    TheMouse wrote:
    Shivsky wrote:

     

    For some reason I doubt many 6 year olds are class A players


    Of course not ... but my post clearly indicated that Class A was just a top of the mind suggestion for a category name and did not imply USCF Class A.

  • #19

    Again, though, it highlights the potential for confusion -- especially among your students.  If you're going to set up an independent rating or ranking/class system I'd suggest staying away from replicating established conventions.

  • #20

    I think a ladder is really the ideal solution here as well.  It's easier to organize than a tournament, conducive to people coming and going, and will give you precisely the kind of ranking you're after.

Online Now