ratings for new players

[tennisfish]

what would be the best way to establish a baseline rating system for schoolchildren who range from ages 6-12.would one start with a swiss type competition and assign arbitrary ratings based on the scores with 1500 as both the average and median?these kids are not really ready for uscf tourneys but i am willing to try to do the calculations based on the elo formulas.i am running a school program for beginners and I also realize that a 10 year old may very well be able to compete with a 12 year old. I welcome any ideas.

[swbobcat]

First I want to thank you for your post. From my own perspective I think your proposed idea of basing it on a rating of 1500 is an excellent idea. I think in order to be of real value it needs to be standards based. Right now in the USCF there is a huge problem with ratings DEFLATION because the vast majority of scholastic players are highly UNDER RATED -- indeed many "scholastic" ratings are based upon such intangibles as AGE and GRADE of the child. Add to this mix that many scholastic chess programs employ rated experts and above to coach said children and you have a kid who in reality has a real rating strength of 1100 but is rated at 800.

Just out of shear curiosity -- I know what they say about CATS and Curiosity -- what is the exact purpose of your rating system?!? Is the purpose for intraschool competition (within the school proper and to have FUN) or is it to prepare them for eventual USCF Scholastic competition?!? Big difference. Me?!? While I STRONGLY support the idea of using chess as TOOL to teach things such as analysis, concentration, etc. which will help boost things such as Science and Math scores -- and doing it in a FUN and creative way, I think there is way too much effort being spent on making scholastic chess a competitive "sport". I know of no other "sport" where children play adults, nor do I know of any other "sport" where children get professional coaching, and adults do not. There are many OLDER ADULTS who resent the intrusion of 8-12 olds into what should be an ADULT competition. Given that many of these "scholastic" players are UNDER RATED, and the subsequent Ratings Deflation that goes with it makes for a potent brew of resentment towards Scholastic players.

I had an experience where I by default had to play an 8 year old girl -- no one else wanted to play her -- in a "friendly" skittles game, but who was receiving Chess Lessons. The club uses standard sized Card Tables and Chairs. The girl had to STAND UP for the entire game because when she sat down in a chair she could barely see over the edge of the table to say nothing of the board. You could tell that the child was receiving lessons, even had her mother not told me. I took the kid out of book from move 1 and stomped her flat. Many ADULTS simply do not want to play against some child who they know is now being coached by some professional. Bringing an 8 y.o. into a club intended for ADULTS bewilders me.

The problem with facing a scholastic player is that no one knows what their true playing strength is; is it the 800 listed, or is it much higher?!? I lost a skittles game to a kid with a USCF rating of 600. With that type of rating I should be able to be stoned, half asleep, and just pushing wood and still beat the person without a sweat. I learned my lesson, and since that time I've adopted a cold hearted approach of stomping them flat as fast as I can, or I'll end up being the stompee. Maybe if Scholastic Players were forced to use a more realist rating system like the one you are proposing at least the child's true playing strength would be known, rather than a game of Russian Roulette.

[swbobcat]

First I want to thank you for your post. From my own perspective I think your proposed idea of basing it on a rating of 1500 is an excellent idea. I think in order to be of real value it needs to be standards based. Right now in the USCF there is a huge problem with ratings DEFLATION because the vast majority of scholastic players are highly UNDER RATED -- indeed many "scholastic" ratings are based upon such intangibles as AGE and GRADE of the child. Add to this mix that many scholastic chess programs employ rated experts and above to coach said children and you have a kid who in reality has a real rating strength of 1100 but is rated at 800.

Just out of shear curiosity -- I know what they say about CATS and Curiosity -- what is the exact purpose of your rating system?!? Is the purpose for intraschool competition (within the school proper and to have FUN) or is it to prepare them for eventual USCF Scholastic competition?!? Big difference. Me?!? While I STRONGLY support the idea of using chess as TOOL to teach things such as analysis, concentration, etc. which will help boost things such as Science and Math scores -- and doing it in a FUN and creative way, I think there is way too much effort being spent on making scholastic chess a competitive "sport". I know of no other "sport" where children play adults, nor do I know of any other "sport" where children get professional coaching, and adults do not. There are many OLDER ADULTS who resent the intrusion of 8-12 olds into what should be an ADULT competition. Given that many of these "scholastic" players are UNDER RATED, and the subsequent Ratings Deflation that goes with it makes for a potent brew of resentment towards Scholastic players.

I had an experience where I by default had to play an 8 year old girl -- no one else wanted to play her -- in a "friendly" skittles game, but who was receiving Chess Lessons. The club uses standard sized Card Tables and Chairs. The girl had to STAND UP for the entire game because when she sat down in a chair she could barely see over the edge of the table to say nothing of the board. You could tell that the child was receiving lessons, even had her mother not told me. I took the kid out of book from move 1 and stomped her flat. Many ADULTS simply do not want to play against some child who they know is now being coached by some professional. Bringing an 8 y.o. into a club intended for ADULTS bewilders me.

The problem with facing a scholastic player is that no one knows what their true playing strength is; is it the 800 listed, or is it much higher?!? I lost a skittles game to a kid with a USCF rating of 600. With that type of rating I should be able to be stoned, half asleep, and just pushing wood and still beat the person without a sweat. I learned my lesson, and since that time I've adopted a cold hearted approach of stomping them flat as fast as I can, or I'll end up being the stompee. Maybe if Scholastic Players were forced to use a more realist rating system like the one you are proposing at least the child's true playing strength would be known, rather than a game of Russian Roulette.

[tennisfish]

My plan is basically twofold: 1. To satisfy a certain mathmatical curiosity on normal distribution and,2.to help organize the kids to compete on a low key level. In this way,I hope to be able to ascertain who needs more attention to my teaching.My style is rather laid back as to instruction yet I do have a format that I hope will lay down a good basis in fundamentals.If the children want to pursue Chess on a higher level the transition,hopefully,will be smooth.Above all, I want them to have fun and to learn .I firmly believe that one enjoys an activity more when they know more and in turn become more successful.I have 10 2 hour sessions at the local library.The average age is about 15 and the time is 6:30 to 8:30 pm.Plus,they seem enthusiastic to play;I keep the lecture part to go not more than 30 mins.It is human nature for them to"get their hands on the wood".Had one session already;have 25 kids!Keep you posted since you show some interest. Tennisfish

[swbobcat]

Greetings Again!!! You raise an interesting question when you state: "1. To satisfy a certain mathmatical curiosity on normal distribution". The question your statement raises is what exactly *is* a "normal distribution"?? There are so many factors that can affect an outcome, and thus skew the distribution. Some of these factors may include things such as 1) Chronological age 2) How long has the child been playing 3) Is the child receiving expert instruction, or is the child actively teaching (ie reading books for their own education) themselves the game, or none of the above 4) How many hours does the child spend on chess each week 5) What is the quality of the competition the child is playing against. 6) etc. The only way you could posibly determine what is a "NORMAL" distribution whould to get as many children of any given age as possible who *claim* to play chess and then have them play each other, then graph the results to determine where the central mass tendency is, which would thus define a "normal distribution" for that age group. The USCF uses a Class "C" rating -- USCF 1400 - 1600 -- as its reference point for an ADULT player who is of unknown strength. You have to remember what the purpose of the rating system is for which is for pairing purposes in a tournament.

The majpor problem with scholastic players is that there is no uniform scholastic rating system in use. Today, as I mentioned, many of these kids are 1) receiving extensive coaching from USCF rated experts and masters in school 2) They are playing in intra-scholastic tournaments in which intangables such as the childs age and grade are used to determine the child's " rating" 3) These "scholastic ratings" are being used for pairing purposes when the child plays in "inter-scholastic" tournaments, as well as when they play in Adult tournaments. No, the only TRUE way you could actually determine a child's initial strength would be to have the child play a USCF Class C player who has a known rating that runs from 1450 on the low side to 1550 on the high side in a series of say 4 games and benchmark off of that. You could also do that with a Class D [1200-1399] player who varies from 1250 on the low side to 1350 on the high side and benchmark off of that. In other words, performance based standards --vs -- intangable based standards [childs age and grade].

I think it is a very, very bad idea of allowing children who have received coaching from rated experts and masters and who have a "scholastic rating" to play in Adult tournaments, as these scholastic players as I pointed out are leading to rating deflation. It would not be so bad if the rating points kept cycling within the USCF, however most -- between 95% - 99% if the graph and data I saw are corrrect -- have usually dropped out of the USCF by the age of 19, thus screwing up Adult ratings. The USCF did its ADULT members no favor when it started "DUAL RATING" Game in 30 minutes to Game in 60 minutes -- skittles time controls really -- as being "Regularly Rated". As was pointed out the USCF did this so scholastic players who have a limited attention span, could obtain a USCF Regular Rating. In reality G/30 - G/60 events are too fast a time control for most Adults so the USCF created a NEW rating system called "Quick Chess" that covers time controls from G/60 to G/5. G/30- G/60 tournaments have become a very popular tournament format that tends to be heavily attended by scholastic players; any Adults who play in these events however usually tend to take a performance hit in their "Regular" rating -- the "GOLD Standard" rating which further exacerbates the ratings deflation. While the USCF can now post the results of any given tournament within 24 hours of its conclusion, and even calculate different ratings for different sections within the same tournament, it steadfastly REFUSES to allow ADULTS to choose if they want to be dual rated or Quick Chess rated ONLY in these G/30-G/60 tournaments. The USCF's almost unstated, but strongly hinted at, reason being is that too many Adults might choose to be ONLY Quick Chess Rated, thus deprive "Johnny" or "Joan" a REGULAR USCF Rating. Thus this whole thing about "Ratings" becomes an important Question.

You are a bravier man than I Tennisfish to hold 10 2 hour sessions with a bunch of 15 y.o. Is that 10 2 hr session per week or per month?? Regardless I cringe at the thought. No I think it is a very good idea to use 1500 as an inital benchmark rating. I might modify that however by adding "instruction points" to the childs base rating, thus use 1500 as a base rating with NO instruction, and maybe up to 10 points additional rating points for every year the child who continues to recive instruction based upon the results of periodic "chess problem exams". Indeed this is CLASSIC education where we measure progress through examination. You would expect children in your program who score high in solving chess problems to be statistically better than a child who has received no instruction, and thus more likely to measure the childs true rating vs an artificial rating that is not standards based. I wish you well Tennisfish with your program. If you get a chance it might be fun to see if you were to hold a "tournament" (assuming they have received NO INSTRUCTION) of these various 15 y.o. to see where the central mass tendency would fall and compare that to a chess playing program set to play at say various USCF strengths -- say 1000, 1200, 1300, 1400, 1500, etc. that might be a good measure of YOUR cohort's true initial palying strength. At the end of the year have the child again play the same program at the same strengths and compare the results pre and post instruction. I would suspect that the central mass tendency would skew to the high side as the children of the "club" receive more instrction and peg the TRUE rating of your club. ie pre instruction the club might have a TRUE 1350 rating (I'm pulling that out of the air) compared the the computer. Post instruction the club rating might have a TRUE rating of 1550. If nothing else you might be blazing a new trail, and maybe put and end to ratings deflation within the USCF.

Best Wishes,

Bobcat