Home » Forum Home » General

Topic: scoring
Replies: 4   Views: 72,910   Pages: 1   Last Post: Jan 28, 2017, 8:41 PM by: valleyboy

Search Forum

Back to Topic List Topics: [ Previous | Next ]
Replies: 4   Views: 72,910   Pages: 1  
valleyboy

Posts: 122
Registered: Mar 2, 2005
From: McAllen TX
Age: 56
scoring
Posted: Jan 28, 2017, 4:51 AM

What are the chances of a scoring algorithm in each set not only as 0-0 with a split set but with the determining percentage applied on each win-loss?


rainwolf

Posts: 766
Registered: Apr 12, 2008
From: Singapore
Age: 44
Home page
Re: scoring
Posted: Jan 28, 2017, 11:21 AM

Do you mean to apply a change in ratings on a draw? Or communicate some calculated score in the server message after a draw but no effect on ratings?

valleyboy

Posts: 122
Registered: Mar 2, 2005
From: McAllen TX
Age: 56
Re: scoring
Posted: Jan 28, 2017, 3:35 PM

A change in ratings. Shouldn't a player winning at least one game against another player with a 3-400 higher score be rewarded?

I'm asking if it's possible to apply the differential of awarded points...more for weaker player for taking an entire set, less for the stronger player...to each game in the set.

rainwolf

Posts: 766
Registered: Apr 12, 2008
From: Singapore
Age: 44
Home page
Re: scoring
Posted: Jan 28, 2017, 7:17 PM

That sounds reasonable, but I don't know how to make that fair.

As I understand it, rated games are played in sets because of a (perceived and probable) advantage for player 1. The fix to that advantage is to make players play as player 1 and player 2, and force a ratings change when both games are won/lost.

Given that advantage, if you still wanted to award points for winning 1 game, I think you'd have to differentiate between player 1 and 2, and then begs the question, how and how much do you differentiate?
I don't see an easy answer to that.

At first glance, it seems that the answer would complicate rating calculations quite a bit and I like to keep things as simple as possible.

Another thought that comes to mind is that, with such a rule, I imagine people become less inclined to play lower ranked players because they'd stand to lose more from a draw.
That's something I'd want to avoid.

I believe rating changes already have incorporated that a lower ranked player stands to lose less on a loss than they can gain with a win, and a higher ranked player can win less than they stand to lose by playing a lower ranked player.

In addition, changing the way ratings are calculated now implies, to me at least, that I should reset everyone's ratings because it seems meaningless to use the results of previous calculations as the base for a new formula.

I'm not inclined to change the rating system, I'd need a very convincing argument for that.

valleyboy

Posts: 122
Registered: Mar 2, 2005
From: McAllen TX
Age: 56
Re: scoring
Posted: Jan 28, 2017, 8:41 PM

Yes, I can see it would take a quite complicated algorithm. And don't look at me...I had to look up how to spell it.

Replies: 4   Views: 72,910   Pages: 1  
Back to Topic List
Topics: [ Previous | Next ]


Powered by Jive Software