Whats the logic behind having scholastic ratings, as opposed to just leaving the status quo as having children’s rating be the same as an adult rating?
Does it somehow water down the rating system?
Someone also mentioned a ratings floor. Although I’m not sure how it could be implimented. If you had an abitrary floor of say 1100, then you would have a hard time paring players rated under 1100 (which presumable would be considered <1100/Unrated).
It wouldn’t be very valuable to pair someone that just learned to moves a couple months ago to another individual thats just barely below the ratings floor threshhold.
Someone posted in a different thread maybe the USCF have ratings floors. Like their is suppose to be some sort of minimum level of proficiancy before getting a rating. (At least, thats the impression I got from the post.)
Seems kinda subjective, especially in light of the massive amount of scholastic players. The other poster gave I think 1100 as a ratings floor.
If we had a floor of 1100, then what do you do with all the players that are sub 1100? Are they considered unrated or what?
I believe you may have the wrong idea about what rating floors are. Here is the short non-detailed version:
Rating floors are calculated individually for each player after they receive their established rating (I seem to recall that comes at the 25 game mark).
The individual floor is calculate by replacing the last two digits of a player’s current rating with zeros (00) and then subtracting 200 from that result. So, a player with an established rating of 975 has the 7 and 5 replaced with zeros giving 900. Now 900-200 = 700, the new rating floor for that individual.
With lower ratings there are one or two benchmarks that once an individual’s rating goes above that point then their rating can not go below that mark again ( I can’t recall what they are at the moment, but I want to say 1000 is one of those benchmarks).
This application of floors applies to all players not just scholastic players.
Ahh, ok. Thanks for the info. I definately had the wrong impression of what a ratings floor was.
I’m trying to find that particlar post again and re-read it though. Can’t remember which thread it was in at the moment.
Mike Nolan may correct me but I thought that the rating floors Tim described were only for floors of 1400 and up. If you’ve never reached 1600 then your floor is 100.
Ok, I found the post I was looking at earlier. I missread it somewhat.
The poster was talking about scholastic chess in Oklahoma, and if kids got proficiant enough for an 1100 rating, they’d give the kid a membership to USCF.
In the same post, he was debating the merits of having seperate rating systems for scholastic and adult chess.
Here is a shortened version of his post (skipping the stuff not related to this thread).
Not sure where I got the impression he was talking about a ratings floor. But probably read that in some other non-related thread and got them mixed up with each other.
The minimum peak rating-based floor is 1400 and the maximum is 2100. Players with 300 or more games as a master qualify for an Original Life Master floor of 2200.
In theory if someone won $2000 or more as an under 1200 prize they could get assigned a 1200 money floor, but I don’t think anyone’s ever held such an event.
Players with very low ratings have a floor based on the number of events and wins they have, up to 150.
That’s a new policy, isn’t it Mike? I remember some discussion about it several months ago, but I don’t remember any policy being based on it. Can you explain in a little more detail for those of us who thought like Mr. Wiewel that floors went straight from 1400 to 100?
The change from an absolute floor of 100 to one of up to 150 is a recent, and not particularly important, change. As I recall, for each lifetime win, a player’s absolute floor is raised by 4 points (but to not more than 150). Draws are good for 2 points each.
That policy change was approved by the Board in May and went into effect in June:
EB 08-42: Move to change the 100 rating floor as follows: The floor will be increased by 4 points for each win, 2 points for each draw, and one point for each tournament (3 games minimum) with no resulting floor to exceed 150. Passed 6-0.
This change was made in response to concerns that ratings were piling up at the 100 floor, and that players were becoming discouraged because they were stuck at this floor. The original proposal didn’t have the 1-point rise per tournament, though.
I was one of those proposing such a change. Then Mike Nolan came through with some data strongly suggesting that players were discouraged not because they had a 100 floor, but because they lost all their games. At that point I became lukewarm, at best, to my own proposal. And when the proviso was added that even losses would raise a player’s absolute floor, I became opposed. The distortion caused to the rating system is negligible, but it opens the door in a bad way.
Any rating which came out below 100 was set at 100. So a lot of players (young scholastic players, mostly) had ratings of 100. Many fewer had ratings of 101, 102, etc. So there was no way to tell whether player X or player Y was better, since both were 100.
The change assures that, as soon as a player wins or draws a game, he will no longer be 100.
That link is to a graph that shows the number of players who had updated ratings under 200 in the August 2008 through January 2009 ratings lists as well as those who were in the August 2007 through January 2008 ratings lists.
As you can see, a year ago we still had a very large number of players at 100. We now have smaller blocks of players with ratings from 100 through about 110 but by 120 the two graphs are nearly identical.
This is not surprising. One way to get from 100 to 120 is by winning (over the course of several tournaments) 5 games. (5 wins * 4 points each = 20 points.)
By the time you’ve won 5 games, you’ve probably improved enough to have a rating well over 120.
I think Mike and Tim are confused. The OP talking about raising the existing Absolute Floor from 0100 to 1100.
To answer his question, the way the Absolute Floor works is: if your computed rating is lower than the floor, it is immediately raised to the floor.
To answer a question I don’t think he asked - there is ZERO chance of the Absolute Floor being raised to 1100.
alas, there will always be foolish tinkering with floors - witness the silly change to modify the 0100 floor by giving rating points for winning/drawing games - even if the player’s computed rating is below 0100.
Not necessarily! If that were the case, then no one would have proposed this silly tweak.
What is true is that the difference between 0100 and 0120 is down in the noise - it’s totally insignificant. So…“why bother?” you might ask. Indeed. Excellent question. I don’t have an answer to that one.
Back when some people thought some players were staying away because they were discouraged by their 100 rating, it seemed to make a little sense, at least.
But once Mike Nolan discovered that players were discouraged because they lost all their games rather than because their ratings stayed at 100, that pretty much took the steam out of the argument.
There remains, however, one possibly valid (though small) argument. When pairings are made, instead of having 50 players at 100, now you’d have maybe 20 at 100, and the rest scattered between 101 and 120. Gives the wallchart a bit more organized appearance.
