"What argument? You haven't defined your system. Define it, I'll rip it to shreds, you ignore it. It'll be fun"

The aim of a rating or points system should to promote best quality play, and discourage quantity-style play; the points system was introduced because players didn't care to play to a standard if they were loosing a game. Therefore the best reward system is to reward playing well and penalise playing badly. This means that the best reward system to use is a system that accurately measures a players skill, because then they are encouraged only to play well, which is what we want. Ghost-Rating is the best system we have to measure player skill. Therefore, Ghost-Rating is the best to encourage the right sort of play.

I think both systems are useful.

As I made clear, "The simple answer is that I don't care about replacing the existing system."

Kestas, I know you're dead set against changing the rating system. That's fine, I've accommodated that to my own satisfaction. What I'm not going to do is enter a long discussion where you give a whole load of "what ifs..." about a rating system. I've tried modifying the system in certain ways before, because that provides against your fears as to what it might do. I've tried suggesting a lowered starting position, tried suggesting a "chance" term, I've given up trying.

You think that "There's no real way to express how good a player really is in a single number." Wrong. That number is simply how likely they are to win a game. You think that the criticisms of points are "the same tired arguments". Wrong. They are the same arguments against the same system, just one year on; nothing has changed. You complain that I don't "provide a specification" for my system, but in April I posted this:

http://forum.webdiplomacy.net/viewtopic.php?f=23&t=135

of course, by this time there was a ratings forum there to split them from things you were interested in. You admit that the only reason for changing the system would be to stop discussions about them. That is why I gave up. You have set the only challenge I am not convinced a rating system could ever surmount. I'm not going to try again.

I might as well throw in my 2 cents too. I'd say most people here can see that neither system is perfect but they both have different advantages that complement each other quite well. That's why it would be so nice to see them both listed on peoples profiles. I'd say most people here value both systems but I've played with people who only care about D points and others who only care about G rating so both are clearly valid to different people.

As a casual gamer who hates CDs I highly value the fact that I can use Kestas system to play games with people who are unlikely to CD. My competitive side though makes me value peoples win/draw and survive/loss ratios and ghost ratings too though.

Either way both Kestasjk and TheGhostmaker have put a lot of work into this so I'd just like to thank you both for the effort you've put in making everyones gaming more fun.

Stripy +1 Well said. Having both systems is great. Thanks to you both for your hard work.

I don't think so. The points system is, in my view, just inferior. That's fine, it wasn't meant to be the last word in rating systems, but I don't see any reason to persist with it when we have an alternative that rewards nothing other than just playing well.

Hey ho.

>Therefore the best reward system is to reward playing well and penalise playing badly.
>This means that the best reward system to use is a system that accurately measures a
>players skill, because then they are encouraged only to play well, which is what we want
I think I addressed this here: http://forum.webdiplomacy.net/viewtopic.php?p=288#p288
There are factors which need real consideration other than skill, as I said then "If this was just a ranking system I wouldn't care; however the majority want to rank themselves would be fine with me, I'm not going to be #1 in any ranking scheme. The problem is the ranking system is just a part of it, and it's closely tied up with things I do care about which keep the site manageable."

Your goals for the Elo system are different to the goals for the points system, this is necessarily so and you have clearly stated your goals and I've stated mine, and they conflict. Yet you keep coming back with "the points system is flawed" based on *your* requirements, and when I point this out you either say "I don't want it to be a replacement system" (then why point out that the existing system is flawed?) or "it can be a replacement system, it has a problem with this or that but it's still doable" (but how then? ratings going below the starting value is no trivial issue, and neither is a lack of inflation (not that your system actually lacks inflation in practice))

I hadn't seen the specification you posted, and I'll have to simulate it some time to see what properties it has, but by your own admission it doesn't meet the criteria the points system was designed to meet, and it's the same system which you dropped the argument on.

Either it is a points system replacement, in which case you've left a /lot/ of questions about points system functionality unanswered and you've got a lot of "what ifs" to answer, or it isn't a points system replacement, in which case why are you calling the points system "flawed" based on your separate league system requirements?

How about
wins divided by defeats multiplied by the sum of draws and survives= R
R multiplied by the average R of previous opponents.
that was really taking all the stats and combining them in a random way.. haha but it seems to work other than if you don't have a win its always zero

"By your own admission it doesn't meet the criteria the points system was designed to meet"

I admit only that it allows ratings going below the starting value.

"not that your system actually lacks inflation in practice"

I beg your pardon? That is just wrong. Every time I have checked, for instance, 230 is, to within a few places, the boundary for the 1st percentile.

I don't just think that the points system is flawed in that it isn't accurate, I think it is actively damaging. Sure, its better than no system, but, if it is to have any influence, people must want to gain points. In points, picking the right players is vitally important if you want to gain points. I have not played to gain points, have not done this, and that is why I have relatively few points. In Ghost-Rating, the players who you play are irrelevant.

I've re-read your post and don't see a response to this very simple argument:

P1: We want to encourage high quality play
IC: Therefore, the quality of play should be what determines your rating
P2: Ghost-Rating measures skill more accurately than points.
P3: Skill and quality of play are equivalent
C: Ghost-Rating is superior to points in terms of encouraging play habits.

As I see it, having some rating system is of enormous value, even a bad one, which is why points have worked. It is not because points are a good system.


I think that Ghost-Rating would be a better system for the site. I don't think I will convince you of the fact without demonstrating it by running a site on it. I don't wish to set up a forked site to do this. That is why I chose not to respond fully to your objections. Simply put, it is a trouble I do not wish to go to.

denis: what would that *mean* though? For a rating system to really work, the number has to have some tangible meaning.

In practise, that meaning is that you should be able to predict the expected (i.e. average) outcome of a game from knowing the ratings of the players.

Not to mention that these discussions always end up getting too acrimonious for my liking.

that what i said a random compliling of the stats it was a random suggestion.