CTP Ratings (03-FEB-2001)

[quinns]

Franses,

Obviously, I used an extreme example to make a point. No, I don't believe there are people that would intentionally do that... but it would be possible. I think that it would surface more subtly in reality. Like, "Hey Quinns, how come so and so is the only one rated in that game and he is getting points for defeating nobody that is rated... hmmmm??"... and "Hey Quinns, why is it that so and so's rating always goes up and never goes down even though he's losing in MOST of the games he is in... hmmmm? And on and on... Yes, I DO believe that I would here THAT in the future if we implemented Keygen's first proposal.

Quinns

[St Jon]

I think you over-value the 'Rank' that someone can achieve as opposed to the normal aim of the Game. Sure I could play for 2nd or 3rd place if all I cared about was how high I would climb statistically but I don't, and I am certain that the majority of other Players don't either.

I am in two Games with Birdman and our positions are reversed whilst we are Allies. If all I cared about was my Rank I would have abandoned him in DC rather than Nuke his aggressor and he certainly would not touch me in Lung's Fast. It's the Game that counts and not the points for we, all of us who have been playing eachother for any length of time, know who is good, bad, dreadful or utterlly superb. Whatever the Rankings may say I still reckon Mobius, closely followed by Klair, to be the best Player I have ever played.

Just my viewpoint.

[quinns]

I suppose you're right Jon. I'm getting a little carried away with this. Maybe it's just that I would like to see the rankings be as pure as possible, as I am trying to achieve that in World Tennis Ratings. I'm hoping that this rating system will some day replace the Association of Tennis Players (professional men) and Women's Tennis Association (professional women) rating systems. (By the way, Klair and Mobius ARE at the top of the CTP PBEM rating system, so it's good to see that the system is matching your expectations.)

I've seen very good feed back on this forum (better than the World Tennis Rating forum!) regarding this system. So I guess I started to push too hard to try make it perfect. Sorry everybody. It's just a game for fun.

edit:

So let me ask all rated players this question. Do you all agree with St. Jon above? He says he speaks for most of you. If that is true, then I'm really going to question continuing this effort. Though these ratings are interesting and I don't mind doing it, it is a bit of work to keep all this straight. If you want for this system to continue, let me know. If you agree with Jon's attitude towards the ratings, then also let me know. Maybe I'm wasting my time. Most of you have shown good support for this so I'm a bit baffled by Jon's statements. Just let me know.

Quinns


[This message has been edited by quinns (edited February 07, 2001).]

[Paul]

I think that the ratings are fun and when they have been going for a while they should probably give a reasonable indication of players' strength, but in my opinion it would be an illusion to think that the ratings will ever give a 100% accurate listing of players' strength. That doesn't mean that you should stop doing the ratings or that we don't appreciate the work that you put into this.

What I think Jon is saying is that although we appreciate these rankings, the players here are not the kind of people who would resort to this kind of cheating to get higher rankings. We are in the first place playing our PBEM games to have some fun, not to get to #1 on the rankings at all cost.

Regarding the rating of permanent replacement players, maybe these subs should not be allowed to be rated from the moment they join but only after they have played enough turns to have had an influence on their ranking.

I am a permanent sub in Monkey 1 from turn 97. The game is now at turn 188, so I have been in for almost half of the game. I would like to think that my current rank in that game is as much or even more my work than that of the player before me.

Maybe you could say that a permanent replacement player can become rated if he has played at least half of the turns in that game (or a third? or a fixed number of turns?).
[This message has been edited by Paul (edited February 08, 2001).]

[Solver]

Paul,
I have already told Franses about the accuraccy of the system.

Quinns and Keygen,
I think we have the best sub coverage that we could possibly use. I would completely refuse proposals one and two by Keygen (sorry if it's too harsh), and I think that giving subs the points the way we do it is fair. For reminder of the substitute rules, see posts be Quinns, and also the PBEM Ladders thread.

So far, this should go as it is, but I am sure we will be adding more rules over time.

------------------
Solver - http://www.aok.20m.com

[Franses]

quinns, you asked whether we like the rating. Well, I do. So I would appreciate if you continue. I suppose everyone who is rated is of a similar opionion.

However, discussing the rating system is also fun. Do not take the criticism too serious. People always tend to criticize. If we stop liking it we will probably stop being rated. Perhaps that is a good measure for you whether your work is appreciated or not.

Franses

[TheBirdMan]

I agree with St. Jon.

The game itself do mean a lot - more than the ranking.

But - when this is written - by supporting each other/be ally, it is much easier to protect your position in the game.

But nobody knows what will happen during a game.

Ally now, enemy later. Problem that could lead to a "divorsing" could be pollution. It could be trade. It could be one wage war against anothers old friend.

Or that could be simple gread for an extra rating point when all common enemies are destroyed

So, the rating system indeed does mean something in this game. Only later on. First you have to survive.

[Keygen]

First of all, before I answer to the three points again I should say this.

The rating system came out from the combination of the vision and the hard work of both Solver and Quinns to deliver an amusing measurement for the PBEM players.
We all appreciate your work and we totally support you and the system the way it is.
MarkG has understood its value and had made it standing on top all the time .
But many players are already involved on the ratings plus that it isn't fully tested yet so I would consider it a necessity these players to mark any possible weaknesses and propose probably aditional rules if they judge that they worth of conversation.
It doesn't look right to me to be ready to drop the whole thing Quinns just because of an opposite proposal or preclude it before even the majority state their opinion on the subject Solver!
Of cource the Admin should have the last choice but just let it unfold first.

Regarding my proposals.

1st proposal: I still insist that the possibilities on someone cheating are bare. As Franses said we could find the cheaters. Quinns said in a post something about a rated game where only one player would get rated. Is this possible? I though that only those games where at list three players would agree to get rated would be rated!?!
Anyway the true ability of a player is messured against all the opponents of a rated game regardless whether some of the players are unrated and they should effect the ratings.

2nd proposal: Currently it truly might hurt the system in case many withdrawals occur. Let's wait a little longer and see. Maybe Paul is right when he says that a sub can affect decisively the civ's course but in some cases the former might be the one who could affect decisively the civ's future course. I will bring up the subject from time to time just for discussion.

3rd proposal: Only Quinns have answered .

One last thing.
As I had said in the first post (it's on the previous page) I don't consider them essential but a good subject to be discussed.
If these proposals make Solver or Quinns upset then please ignore them .

[This message has been edited by Keygen (edited February 08, 2001).]

[Solver]

Keygen,
Rating system is never complete, and it always can be modified with new rules and proposals. We want to make it better than it, no doubt.

------------------
Solver - http://www.aok.20m.com

[St Jon]

Quinns, Please do not believe that I dismiss the time and effort that you have put in to the Rating System. I do believe it to be a good thing, I would not be Rated myself otherwise, but they should be a reflection of relative ability rather than a motive in themselves.

I play for fun and to win. Settling for 2nd place is not my style and I would prefer to go out in a blaze of glory at least 'trying' rather than just feeding off the scraps from the rich mans table.

I am in several Games where I have known for many many Turns that my chances were slim to non-existant but that does not mean I don't still work as hard as I can to achieve victory. It is the Game that should make the Ratings, not the other way around, and I think most other Players would agree. The kudos of being No1 when it is unearned is worthless and it would be a pretty shallow human-being who would think otherwise.

[Paul]

Monkey 1 turn 189 powergraph:

1. Swissy - Jamaican
2. Paul - Russian
3. Slamp - English
(4. Max Webster - American)
(5. Robm - Roman)
6. Yeti - Chinese
(7. Duckhunter - German)
(8. Fraze - Scottish)

[Keygen]

I won't adduce any additional explanation or try to persuade farther more.
I just want to reply on two points of Quinns answer.

Originally posted by quinns on 02-08-2001 12:51 PM
Let's use Paul's last "Monkey 1" example to help explain why Keygen's proposal would cause unfairness in the system (though it may not be intentional).

There is no "may" Quinns. There is defenetely no intention at all. For God's sake I am about to start 4 games wich will be rated!

Originally posted by quinns on 02-08-2001 12:51 PM
What happens is that the more games that someone plays, the higher their rating goes. This is not intentional cheating from players, this is just what happens. It is not fair that someone's rating should increase greatly just because they have time to play 10 rated games at once, while another player only has time to play 2 games at a time. Even though those two games could be against much better players than the ten games the other player is participating in.

What!?!?!?!
This is already happening Quinns!!!
That's why I had proposed to a previous post on the first page of this thread that the scores should come from the average of each independed score of every game the player is involved.
Are you jocking me?

Originally posted by quinns on 02-08-2001 12:51 PM
Keygen, the statistical model for this can get quite complex. Basically, if one player's rating goes up, another player's rating MUST GO DOWN by the same exact amount. This is crucial to the integrity of the system.

I hope this has nothing to do with my "intensions" Quinns.


Common man don't give me that crap.
All I wanted to do is to make the system as better as possible according to my crisis on the subject.
Maybe I am totaly wrong but don't try to present me with irrelevant cases which are already existing by the way.

[TheBirdMan]

I might be the one, that are playing most (rated) games and I was in lead. Not because of my many games, but because of my positions in these games.

I was only able to stay on top until DC counted (one turn (119) should be cancelled - I talk to Slamp later on this matter).

I do see Quinns point now. Ratings are only for rated players. And I think this is the right way of calculating.

No player (well I think so) will ever go higher than say 30-40 rating points IF he/she is in more than one game UNLESS he/she was second to God in skill and knowledge AND with unlimited luck.

And none would go lower than 5-10 points UNLESS he/she was in only one game AND totally blank on playing this game AND with no luck what so ever.

"Problem" COULD be the player, that only played in one rated game - IF we WOULD do it to a problem. This player could be the highest (or lowest) ranked player ever, depending of luck/skill in only one game.

Ohh - I think I had better stop what I am doing and go to bed

[quinns]

Thanks Birdman for understanding my points and seeing the reasoning behind them. Your support is crucial here, seeing that you were the one affected the worst by the last rating change.

Keygen... I give up on you. I tried to go to great lengths to kindly explain the reasoning behind why your proposals won't work, and instead of trying to understand them, you take it personally and start to throw insults. You resort to statements like, "Don't give me that crap!" . Well, I won't give you any more CRAP. Your proposals are rejected. No more explanations given.

[St Jon]

Can we cut the personal bo11ocks please!

I play for fun and I am certain that I speak for the vast majority of Players when saying that.

This is a bloody game at the end of the day, ie something we engage in for recreation after undergoing a day of harsh reality, not a qualifier for inclusion in the human gene-pool!

For ****'s sake, both of you, GROW UP!!!!!!!!!!!