Page 4 of 7

Posted: Thu Jun 07, 2007 6:09 am
by Vortan
I assume D means people quitting or surrendering not RIPing or surrendering.

Several things:

Sal - you stats freak :P - I am not for a minute suggesting binning the currently collected data so you could for me have all the stats you want.

Kor - I tend to agree with Sal's stance regards your three proposals. Yes it is ultimately a wargame BUT as history shows there were many winners in WW2 not just the British & Americans. Likewise with WOK we could, and should reward the better of the 'losers' but by all means do this as well as other stats.

Cal - Surely your not fed up with the discussion already - I thought you said we had months to go yet :roll: No but seriously I agree that an end date should be imposed because as my Mrs (TheDragon) is pointing out to me this thread is in danger of becoming very boring. A concensus reached quickly and revised later as required is better that one rambled on about for weeks never mind months :lol:

Posted: Thu Jun 07, 2007 12:46 pm
by Aussie Gaz
The old rating system (that was on Donut's website I think).

1st +0.120
2nd +0.090
3rd +0.060
4th +0.030
5th +0.000
6th -0.030
7th -0.060
8th -0.090
9th -0.120
10th -0.150

+0.030 bonus for a rip (only the first 5 rips in a game counted).

-0.030 for being one of the five.

-0.030 for going M-3.

All scores were cumulative. Everyone started on a score of 0. Good consistant players had a postive rating. Those who were frequently ripped early on had a negative rating.

I think from memory the table showed the number of games played also. (and or the average rating per game).

Gaz

Posted: Thu Jun 07, 2007 1:25 pm
by korexus
The ratings system was a good start. The RIP bonuses are what caused most of the issues. I'd suggest leaving them out.

Of course, these still would have a tendency to grow over time. I have an unusual suggestion here: Instead of resetting evryone to rating 1.000 at the end of a year, average their current rating with 1. This regression will mean that past games still count, but the longer ago they were, the less they influence your current rating.


korexus.

Posted: Thu Jun 07, 2007 1:40 pm
by Saladin
It shows that i wasn't paying too much attention to the ratings back then because i thought they would be something like a score divided by the number of games you played. But i see it's a running total as well.

I would personally prefer a running total for the yearly score with whole numbers and no negative values. The end effect is the same but it just looks better to have a score of 310 instead of 0.61.

Posted: Thu Jun 07, 2007 3:15 pm
by Vortan
Once again I agree with Saladin ... this will have to stop he might start to get a superiority complex.

Whole, positive numbers are easier to understand if your not a MATHS TEACHER cough*korexus*cough and are also much easier to explain to the new player who wonders how such numbers were arrived at. Again I say, why complicate it unnecessarily? Sure complex is good for those who can cope with it but at the end of the day knowing that you will get points for finishing in the top 5 will, in my opinion, spur some people (self included) to try a little harder simply to secure some of those valuable little points.

Shocking a statistic as this may seem but 10% of the British populace are innumerate! Thus a complicated formula based system would, should they manage to find us, scare the pants of them. Lets keep it simple.

Posted: Thu Jun 07, 2007 3:44 pm
by Dragonette
I think the scores AG shown were good but i agree that a rip score and all the others make it more complicated. So maybe we could have that system but without the extras.

Then why don't we have two scores one being the complete score of the player [ all the points theyve earned] which is their current score.

The other Could be a yearly score so at the end of a full year we can see who got what[ top 5 for example] At the end of the year that score gets added to your first and the year score starts again.

Dragonette

Posted: Thu Jun 07, 2007 6:00 pm
by Vortan
Just a quick aside - kor :?: - did you get permission or general agreement to go all orange and green :?:

For those wondering what on earth I am talking about see the administrator and moderator tags at the bottom of the forum page.

Have you nothing more constructive to do :?: :twisted: :wink:

:shock: What do you mean '&*!&**!!$*' there is no need for that :shock:

Posted: Fri Jun 08, 2007 12:28 am
by korexus
Oh dear. Points which actually need answering, and I will surely forget them if I leave them till tomorrow...

For those of you who didn't see my other post(s) please be aware that I am *well over the limit*...

anyway.

Saladin: You never pay attention, we're used to it! :P Does it really make a difference if you go up 0.03 points or 2 points after a win, so long as you go up consistently? (From a new player point of view, smaller numbers are better as they seem more attainable...)

Vortan: 67% of British 13 year olds cannot recoginise a triangle upside down. Tht does not mean I will stop using triangles when they are useful; maths teacher or no. At the end of the day, it doesn't matter if users understand the forumula so long as achievement is recognised. Fairness is surely preferable to simplicty in this case?

Dragonette: I think this is pretty close to what I suggested, are there any changes you'd like to highlight, or shall I count you as agreeing?

Vortan, second: The colours were actually set to come up all along, but for some people (including me!) they didn't for a long time. I don't know exactly what I did to fix it, which is why I haven't claimed credit for it, but I figured it was better than the old bold system, so I woudn't try to undo it...



Chris.

Posted: Fri Jun 08, 2007 6:21 am
by Vortan
You drunkard :!: :P

Can't we compromise and have FAIR and SIMPLE?

If you want to add extra points for RIP's then fair enough but lets at least work in whole numbers. The need to work in small decimals is something I fail to understand. Simply multiply the results by 10, 100 or whatever and we have whole numbers. But I don't need to tell you that because you should be sober by the time you read this.

Going back to the score system then:

1st 50, Joint 20, 2nd 15, 3rd 5, 4th 3, 5th 2 with 6th-10th getting 0

+1 bonus for each RIP acheived (to your stated max of 5)
Points halved in event of 5 Quits.
Surrender button option to secure finish position.
Quitting gets nothing.

Would it really be more justified if it were:

1st 0.05, 2nd 0.015, 3rd 0.005, 4th 0.003, 5th 0.002

I mean ... come on already

Posted: Fri Jun 08, 2007 6:30 am
by Dragonette
Vortan
WEve already thought about coming to agreement on a idea then you stir it up well don't. I preffer the other ideas.

Count me in korexus.By the way my mum and dad just put two 'upside' down triangles in front of me and i got it right.If you came into my and put 'upside' down triangles in front of us then we'd get it right but then its proberbly something to do with the fact that were all working at all level 7 or level 8.

Dragonette

Posted: Fri Jun 08, 2007 6:41 am
by Vortan
Oh I see... So the Scholars have decided have they ... and the rest of us can go hang. Is that it?

Posted: Fri Jun 08, 2007 6:49 am
by Dragonette
No i just think the other ideas are better.Its not that were deciding without you.

Dragonette

Posted: Fri Jun 08, 2007 11:00 am
by korexus
What on earth are you doing, checking WoK at 6:30 in the morning??!!!

Points for RIP bonuses is a bad idea. I only expect to say that once. Anyone that doesn't know why hasn't read the first set of threads on ratings...

Vortan, I think you missed my point. I was merely explaining to Sal that decimals and whole numbers make no difference to the counting method, precisely because we can multiply everything by an arbitrary constant. This is then what you're saying to me too! :wink:

As for the triangles, to defend our thickos slightly, the question gave several triangles and various other shapes. The had to circle all the triangles and 67% of the kids missed at least the upside down one...

Not much of a defense, but there you go. It keeps me in a job anyway!

Posted: Fri Jun 08, 2007 11:03 am
by Calidus
OK, since I am a little dense, and I really don't have the time or desire to go back and read this whole post, someone please layout the current proposal for the rating systems. This way us lazy players don't have to sort through the good and bad ideas.

Thanks
~Calidus

Posted: Fri Jun 08, 2007 11:03 am
by Calidus
OK, since I am a little dense, and I really don't have the time or desire to go back and read this whole post, someone please layout the current proposal for the rating systems. This way us lazy players don't have to sort through the good and bad ideas.

Thanks
~Calidus

Posted: Fri Jun 08, 2007 1:04 pm
by Vortan
Going back to the score system then:

1st 50, Joint 20, 2nd 15, 3rd 5, 4th 3, 5th 2 with 6th-10th getting 0

Points halved in event of 5 Quits.
Surrender button option to secure finish position.
Quitting gets nothing.


The above from post timed 6.21am Fri 8th June - about five posts up you lazy individual. And NO I did not want to be up at that time! Blame Dragonette she will insist on going to school.

I think we may be reaching a concensus :roll: Does anyone actually object to this because if not can we finalise this because we now seem to be discussing it just for the sake of it and frankly I can think of other things to do ie PLAYING WOK!

Posted: Fri Jun 08, 2007 1:05 pm
by Vortan
Oh ... and you notice how he admits to being dense AFTER getting us to join C.o.N. :lol:

Posted: Fri Jun 08, 2007 1:16 pm
by Saladin
I think the difference between winner(s) and the runners up should be much bigger. It should really show that you have won a game.

A split win should give you more points than 3 third place finishes.

So

Code: Select all

Split win 100
3rd 30
4th 20
5th 10
6th 5
7th 3
8th 2
9th 1
10th 0

If there is a solo win then:

Code: Select all

Solo win 250
2nd 50
3rd 30
4th 20
5th 10
6th 5
7th 3
8th 2
9th 1
10th 0
the rest i agree with Vortan:
Points halved in event of 5 Quits.
Surrender button option to secure finish position.
Quitting gets nothing.
[/quote]

Posted: Fri Jun 08, 2007 1:52 pm
by Saladin
Saladin wrote:I think the difference between winner(s) and the runners up should be much bigger. It should really show that you have won a game.

A split win should give you more points than 3 third place finishes.

So

Code: Select all

Split win 100
3rd 30
4th 20
5th 10
6th 5
7th 3
8th 2
9th 1
10th 0

If there is a solo win then:

Code: Select all

Solo win 250
2nd 50
3rd 30
4th 20
5th 10
6th 5
7th 3
8th 2
9th 1
10th 0
the rest i agree with Vortan:
Points halved in event of 5 Quits.
Surrender button option to secure finish position.
Quitting gets nothing.

Posted: Fri Jun 08, 2007 1:54 pm
by Saladin
Chris you might want to look at this. I spotted a typo in my post. Edited it and instead of replacing the original post it was added as a new post.