03 May 08
This is more or less a copy and paste from my suggestion in the only chess forum
I have often seen players around 1800 for example who, either through massive timeouts or other reasons, drop 400 or 500 points, and then regain all those points later. During that time, players of 1300 and 1400 strength play them expecting an even game, and unfairly play an 1800 player with a 1400 rating. I think rating floors like the USCF's would be a good idea to remedy this.
15 May 08
Originally posted by PhlabibitYour age multiplied by 13.876 divided by 1.763, Then add the age of your granny, take away the age of your oldest cut toenail, and the number of hairs on the palm of your hand and then divide the answer by the number of hairs on your left bum cheek.
What is your decided formula? Top rating minus... ?
Average rating minus.. ?
P-
15 May 08
5 edits
Originally posted by PhlabibitCaissad4 Said:
What is your decided formula? Top rating minus... ?
Average rating minus.. ?
P-
I like the way it was suggested in USCF: Create a permanent rating floor by rounding highest rating down (ex: a 2050 rated would become 2000 then subtract 200 points, making it impossible for a 2000 player to ever play in banded tourneys below 1800. If that player ever reaches 2100 his permanent floor would become 1900.
I think this seems like a very good scale to use? V R
How to determine what a users rating is could be done by the graph perhaps? The rating that is most constant. I truly don't even know what mine would be as I go between 1400's and 1800. I am sure my graph would show where I am mostly at though.
Thank You Russ, however you decide to do it 🙂
15 May 08
Originally posted by Very RustyMy highest rating ever was around 1570. If my floor was 1370, that would be just fine.
Caissad4 Said:
I like the way it was suggested in USCF: Create a permanent rating floor by rounding highest rating down (ex: a 2050 rated would become 2000 then subtract 200 points, making it impossible for a 2000 player to ever play in banded tourneys below 1800. If that player ever reaches 2100 his permanent floor would become 1900.
I think this see ...[text shortened]... raph would show where I am mostly at though.
Thank You Russ, however you decide to do it 🙂
Floors would also help when someone gets timed out in a hundred games. You were playing him and doing well, next their rating crashes to near nothing and you get jack!
Will also be interesting... new graphs would flat line, rather than "V" dive to near zero.
P-
15 May 08
1 edit
Originally posted by PhlabibitI think mine at 1800 is inflated from game load, but I could live with the 1600 floor rating.
My highest rating ever was around 1570. If my floor was 1370, that would be just fine.
Floors would also help when someone gets timed out in a hundred games. You were playing him and doing well, next their rating crashes to near nothing and you get jack!
Will also be interesting... new graphs would flat line, rather than "V" dive to near zero.
P-
This should help solve the sandbagging, where the high rated players, lowers their rating to play in a lower rated tourney.
I guess the next problem will be people purposely keeping their ratings low...But the bigger part of the problem will certainly be solved. 🙂
I believe if done the USCF way it is rounded off to the closest 100, if I under stand it correctly. Then -200 points ! Someone feel free to correct me on this if I am wrong.
16 May 08
Originally posted by Very RustyI thought USCF did not set the floor until X games were played at a certain rating class, but I could be wrong.
I think mine at 1800 is inflated from game load, but I could live with the 1600 floor rating.
This should help solve the sandbagging, where the high rated players, lowers their rating to play in a lower rated tourney.
I guess the next problem will be people purposely keeping their ratings low...But the bigger part of the problem will certainly be so ...[text shortened]... er stand it correctly. Then -200 points ! Someone feel free to correct me on this if I am wrong.
You're correct that USCF rounds down to the nearest hundred and then subtracts some set value of points [I think it is 200, as you and caissad4 have said, but I'm not 100% sure].
- Joined
- 07 Jun 05
- Moves
- 5301
16 May 08
Originally posted by RussCan someone explain to me how this will not lead to grade inflation?
This is coming very soon. (Next few days)
-Russ
To exaggerate, if someone rated 2401, and playing 100 games resigns everything, then 100 players get a rating boost by winning against a 2200 (if the system above is used) rated player. But no-one loses any points. Overall, more points get added to the system - giving inflation.
Why I care:
I try hard to play my best, and play (veerrry) slowly. Even if the inflation is only 50 points per year, I'll struggle to keep up!
Cheers,
Gezza
16 May 08
Originally posted by gezzaIn the grand scope of things, 100 players getting an extra 10 to 40 points now and again shouldn't mess things up too much. It will trickle down, and perhaps average RHP ratings may go up 1 or 3 points.
Can someone explain to me how this will not lead to grade inflation?
To exaggerate, if someone rated 2401, and playing 100 games resigns everything, then 100 players get a rating boost by winning against a 2200 (if the system above is used) rated player. But no-one loses any points. Overall, more points get added to the system - giving inflation.
Why ...[text shortened]... Even if the inflation is only 50 points per year, I'll struggle to keep up!
Cheers,
Gezza
What goes up, will come down.
P-