Average opponent rating

Average opponent rating

Site Ideas

Cookies help us deliver our Services. By using our Services or clicking I agree, you agree to our use of cookies. Learn More.

K
Chess Warrior

Riga

Joined
05 Jan 05
Moves
24932
25 Mar 08

I think it would be not bad idea to include information about average opponent rating in players profile.

R
The Rams

Joined
04 Sep 06
Moves
13491
26 Mar 08

Seconded

R
The Blues Boy

Joined
18 Jan 08
Moves
9229
26 Mar 08

good idea

Already mated

Omaha, Nebraska, USA

Joined
04 Jul 06
Moves
1115070
26 Mar 08

good suggestion

n
The Conductor

With the band

Joined
14 Jun 07
Moves
41110
27 Mar 08
1 edit

Originally posted by Korch
I think it would be not bad idea to include information about average opponent rating in players profile.
Thread 87048

I think this is a good idea.

np

a

THORNINYOURSIDE

Joined
04 Sep 04
Moves
245624
27 Mar 08

Originally posted by Korch
I think it would be not bad idea to include information about average opponent rating in players profile.
Why?

It is an irrelevant figure which can be horribly skewed.

Lets assume that you play 25 players rated at 1200 and 25 rated at 2,000. This will give an average opponent rating of 1,600. Totally at odds with what you have played.

Lets say you play some "dodgy" players whose ratings have dropped from 1900 to 1200 (timeouts etc). Have you played a 1900 rater or a 1200 rater?

Just my 2p worth

k

Joined
31 Oct 03
Moves
17163
27 Mar 08

Originally posted by adramforall
Why?

It is an irrelevant figure which can be horribly skewed.

Lets assume that you play 25 players rated at 1200 and 25 rated at 2,000. This will give an average opponent rating of 1,600. Totally at odds with what you have played.

Lets say you play some "dodgy" players whose ratings have dropped from 1900 to 1200 (timeouts etc). Have you played a 1900 rater or a 1200 rater?

Just my 2p worth
We do not have to have this necessarily, I agree.

K
Chess Warrior

Riga

Joined
05 Jan 05
Moves
24932
27 Mar 08

Originally posted by adramforall
Why?

It is an irrelevant figure which can be horribly skewed.

Lets assume that you play 25 players rated at 1200 and 25 rated at 2,000. This will give an average opponent rating of 1,600. Totally at odds with what you have played.

Lets say you play some "dodgy" players whose ratings have dropped from 1900 to 1200 (timeouts etc). Have you played a 1900 rater or a 1200 rater?

Just my 2p worth
Similar objections can be addressed to rating system at all.

k

Joined
31 Oct 03
Moves
17163
27 Mar 08
1 edit

Originally posted by Korch
Similar objections can be addressed to rating system at all.
Let's add the "best win against" (won against highest rated player) stat too then.

Also, worst loss---lost against lowest rated playa.

K
Chess Warrior

Riga

Joined
05 Jan 05
Moves
24932
27 Mar 08

Originally posted by kenan
Let's add the "best win against" (won against highest rated player) stat too then.

Also, worst loss---lost against lowest rated playa.
I agree - its good idea.

SS

Joined
15 Aug 05
Moves
96595
27 Mar 08
1 edit

A

Amsterdam

Joined
04 Feb 06
Moves
48636
27 Mar 08

I do support some kind of average rating, but to use it at setting up clan challenges:

Thread 82439

R
The Rams

Joined
04 Sep 06
Moves
13491
29 Mar 08

Implement it, I'd say at least 75% support it.

Joined
07 Aug 06
Moves
117505
11 Apr 08

how about the moves per month as a moving base over the previous month from the current date

a more valuable assessment of an opponents status of activity

in essence an average over a month for say thirty days

whereas at the moment, it only approaches this at the end of the month

and could be a little irrelevant at the beginning




and i still like the idea of stabilising 'ups and downs' via a tempering of scores based on moves per game, or some such....

perhaps timeouts could also be a lesser drop in ones score, i've seen people leave as a consequence of a forced absence, then find a plummetted score as all being just a little too hard

then those that remain on the site adversely affect normally scored individuals - further enhancing the fluctuations

what about using a similar ranking as say tournament entry score as the relevant base to a game / game score




just sitting thinking pondering......

P

Joined
06 May 05
Moves
9174
11 Apr 08

It would be possible to be skewed, but could be an interesting statistic.

I personally would like to be able to go to a game I played and see with the game the rating both me and my opponent were at when the game ended.

I think this might be the case in my opponent, but my rating always shows my current rating.