I always found it interesting that people choose to play in tournaments to "test" and "perfect" their strategies.  You're certainly not alone, BD.  But I've always considered that Tournament Ratings are just as meaningful and useful as any other rating on this site.  After all, suppose I want to choose a strong partner to play with in a tournament. I would look at their tournament rating, right?  I mean, why bother to have it in the first place?

I figure if you want to cut your teeth on a board, play private (unrated) games.  This is also an argument for the ability to invite the Public to a Private game, or put another way, make a Public game Unrated.   I'm pretty sure the only reason this is not an offering is that some (including tom) think it's too confusing for the general membership, but really, it makes a lot of sense to have it when practically every game the casual player can find on this site is rated.

So.. "If" WG were to have a composite rating system, I would be in favor of having as inclusive a system as possible.  My vote is still for including tournament ratings; TR is much easier to include than CP because TR and GR are computed similarly.  Everyone agrees that CP is important though, so it's worth it to find a way to work it into the system, and though logs are a bit weird, I'm ok with that because the whole thing is an artificial construct.

If we can agree on something that fits the "Overall" bill, perhaps tom would be willing to throw up an "experimental" O-Rating column.

To be honest I don't think any of the points are a reflection of anything. I could easily manipulate the system in place to get a higher rank, however I play games I am more likely to enjoy, not win. The point system is kind of fun to have around and might give a general idea of the player's performance, but I reiterate: I don't think it's an accurate reflection of anything.

+1 to everything M57 said. 

I also like the general idea of having this composite ranking.  It definitely provides another goal to shoot for, and like that you can approach it in different ways.

Mad Bomber wrote:

Global Rankings favor light use too much......

I don't see how Global Rankings favors light use.  In that they are volatile, I suppose you could play until your GR is high and then stop playing, but what would be the point of that?

CP points are hard to come by....I could have a score of 1500 on a board and not have any CP points......i could play 24 games and have the high score for global rankings

True.  CP points are much harder to come by on popular boards.  10 of my 12 points come from my Assassin board, where I am the top ranked player with less than 1400 points.   With more and more boards going live on the site, there is potential to game that particular ratings system by specializing in less popular boards.

 The idea is to put together a compelling argument for overall bragging rights. A reasonably balanced composite score based on CP, GR, and HR makes  it pretty hard to game the system.  Or, to put it another way - to the degree that players can "game" such a system, they deserve their bragging rights.

If and when WG  has a reasonable way for players to play unranked public games, then I see no reason that Tournament Points shouldn't be thrown into the equation. 

M57 wrote

With more and more boards going live on the site, there is potential to game that particular ratings system by specializing in less popular boards.

In some ways I think this is good, because it will encourage players to try out some of the less popular boards, and is self-correcting because as more players play the less popular boards to try and get Champion points, the boards will become more popular.

Personally, I don't see the need for another score/calculation*.  New players aren't going to care/know/figure out what some crazy calculation based on a slew of data, logs, and exponents means and veteran players aren't going to need some crazy calculation because for the most part veteran players are able to knew who is good and who isn't.

I'd much more see Achievements type of system rather than another score.

* = I see the coolness of Team Rankings, so I would be in favor of that new score.

Yertle wrote:

Personally, I don't see the need for another score/calculation*.  New players aren't going to care/know/figure out what some crazy calculation based on a slew of data, logs, and exponents means and veteran players aren't going to need some crazy calculation because for the most part veteran players are able to knew who is good and who isn't.

I'd much more see Achievements type of system rather than another score.

* = I see the coolness of Team Rankings, so I would be in favor of that new score.

 

this

Yertle wrote:

Personally, I don't see the need for another score/calculation*.  New players aren't going to care/know/figure out what some crazy calculation based on a slew of data, logs, and exponents means and veteran players aren't going to need some crazy calculation because for the most part veteran players are able to knew who is good and who isn't.

I'd much more see Achievements type of system rather than another score.

* = I see the coolness of Team Rankings, so I would be in favor of that new score.

..then go start your own thread .  But seriously, Yertle's point is well-taken and in my mind it brings to the fore a few of the more subtle talking points of this thread.

Correct me if I'm wrong, but originally CP's were originally designed to be the primary system used to identify the top players (though there has never been universal agreement about this) . Now WG has a bunch of disparate scores/ratings that are intended to help us track the top players. The irony is the more types of scores there are, the less easily we are able to determine who the top players are.  So we end up with oxymoronish thoughts:

• We don't need another "score/rating" system ..yet wouldn't it be cool to have team rankings, etc..
• We don't need another "score/rating" system ..yet wouldn't it be cool to have one to rule them all.

Personally, I agree with both of the above statements ..whatever that means.

I see the Team Rankings as completing the current system, not adding another/different way to track the top player.  It fills an existing gap, it doesn't necessarily expand the system...does that make sense?  

Yertle wrote:

I see the Team Rankings as completing the current system, not adding another/different way to track the top player.  It fills an existing gap, it doesn't necessarily expand the system...does that make sense?  

It only completes the current system if you are sure that it is the the last possible (or reasonable) way to quantitatively track player's competitive, skills, abilities and achievements.  The system is potentially scalable, right?

Couldn't we just as easily say that a Composite Ranking completes the system by giving all of its parts context?

I think for a few of us, like 5RocketCars, myself, Vacano (and to a lesser degree Toto) wouldn't mind an integration for tourneys. We get hooked into these damned :) 1v1 fog antfarm round robin tourneys... to do well, you have to watch the histories every turn, and often there are multiple tourneys at a a time with like 5 games at once...

I just don't have any time to chase points. I fill the rest of my games for fun. So other than Toto, there are a few of us that are going to be underestimated if the tourney scores are not calculated... so maybe that is just fine, at your peril of course!

The easy solution is just to have a combined type score in the formula without tourneys, and then run it again to get another score considering tournaments.

My vote too for a team rankings score.  I'd also like to see some kind of composite score (although I'm not sure how it should be calculated), and it should be promoted as "the" score (in bold or larger font,etc.).  I think that would actually help with confusion that newcomers might have with multiple scores.

An achievement system would be even better than an composite score, but achievements are probably a lot more work for Tom to implement well.

M57 wrote:

Yertle wrote:

I see the Team Rankings as completing the current system, not adding another/different way to track the top player.  It fills an existing gap, it doesn't necessarily expand the system...does that make sense?  

It only completes the current system if you are sure that it is the the last possible (or reasonable) way to quantitatively track player's competitive, skills, abilities and achievements.  The system is potentially scalable, right?

Couldn't we just as easily say that a Composite Ranking completes the system by giving all of its parts context?

It completes the current system by filling the existing gap, which is there is not a Ranking system for Team games (other than wins and break-down by wins).  Sure there are other ways to calculate how good a player is, but, again, I would contend that another ranking isn't necessarily going to give much light.  

Players in this thread know quite a bit about the other players in this thread...I know I don't want to play BD in Risky Kong, Gimli in Bomb Factory, Hugh in Five, etc. and pretty much know to watch out for them in other basic games another ranking system isn't really going to give me any more info.  A new player probably isn't going to care what another player's WarGear Ranking ("O-Rating" or whatever), as it's probably not going to really give much information that simply CP or Global Ranking doesn't already give (H-Rating gives some decent info, but probably isn't really understood/comprehended without some investigation).  Players in between can lean either way, already know or don't care/can use the basics.  Again, I just don't see much benefit in another ranking system.

That leaves the new ranking as another "bragging right", in which I would side with work on Achievements, something more simple and gives new players to veteran players some fun goals.

Just my opinion though.

...I know I don't want to play BD in Risky Kong, Gimli in Bomb Factory, Hugh in Five, etc

Wow...I thought the fun of playing games was competing...what a rat race world this has become.  I remember when I last cared about accumulating points - I was a teenager playing Robotron 2084...killed so many of those brainy dudes that shoot electrodes...I think I tallied a gazillion points or something.

That was a bit of a joke...I don't play a couple of those boards much at all and it was meant more as a "definitely watch out for this player in this type of game" than actually avoiding them. 

Gimli wrote:

...

The easy solution is just to have a combined type score in the formula without tourneys, and then run it again to get another score considering tournaments.

 

As Yertle and BD are already saying a composite rating would add confusion, I don't think it's a good idea to add 2 of them. And obviously, as we are looking for a site-wide picture of skills and achievements (as said M57), tournaments have to be a part of it (I even believe the number of trophies should be taken into account). BD's idea ("what should be used is the average of global ranking and T score weighted by number of games played for each, and use this instead of global ranking") could be a good compromise.

And, Yertle, if you think this composite rating is not needed for many players, it doesn't harm neither to have it. I am very convinced, once we would agree on how to calculate it, that it will become THE site reference ranking in most players' mind.

A complete achievements system is something else, and both would be a huge improvment for the site.

Re: Yertle's point about confusion,

Which is more confusing, having a composite number whose algorithm is somewhat complex, or having no agreed upon way to determine who the top player is?

I would point out that the proposed equation (or maybe another one that we have yet to figure out) is no more difficult to understand than the workings of the points and ratings systems. For example, How many players really understand how H-ratings work?  The description of what the H-rating does in the FAQ is all most people need to know. Most probably look at the equation and say, "Yeah, whatever." How many of us really understand how luck stats are calculated? There are a lot of day to day stats and measures out there that we use all the time, whose inner workings we are only vaguely aware of.   As long as a description of what the composite number represents is well-presented, members will understand its intrinsic value. I would note as a caveat that whatever the algorithm is, it needs to be relevant and robust.. We may or may not be there yet.

Re: Thingol's comment,

I'm not good enough to be a "point chaser", but I like to know where I stand, whether or not I'm a competitive player, and who the competition is.  It might be a rat race for those at the top, but for the rest of us, these things are just handy-dandy gauges.

The difference between this "super ranking" and the H-score (and luck stats) is it is being proposed that this be *the* ranking that determines who is the #1 player on the site.  H-score (and luck stats), on the other hand, are interesting statistics for those of us that care about such things.

If we keep this discussion, and any resulting formula, to an "interesting statistic" then I see no real problem, but I really don't think this "super ranking" can or should be the ultimate ranking on the site.  It is more of a Frankenstein statistic that is attempting to put together several poorly understood and possibly poorly defined statistics (and pseudo-statistics) and somehow come up with something better. 

Might be a bit late with this point, but right now the proposed "super ranking" has 4 categories: CP, GR, H-score, and Tournament Ranking. 

It is possible to gain a very high rating in 3 out of the 4 categories (GR, Tournament Ranking and H-score) without playing a lot of different boards.  I would think  the weight of CPs should be higher than the other categories just because you have to play many boards to get somewhere on that stat (as opposed to specializing in a few boards - which I tend to do).

Amidon37 wrote:

..but I really don't think this "super ranking" can or should be the ultimate ranking on the site.  It is more of a Frankenstein statistic that is attempting to put together several poorly understood and possibly poorly defined statistics (and pseudo-statistics) and somehow come up with something better. 

The other rankings may be poorly understood, but I would disagree that they are poorly defined.

All of the "Scores" are psuedo-ish in that they they start at an arbitrary number (1000), and fan out in accordance with an algorithm that scales that number using a multiplier that has  been decided on in arbitrary fashion, but the same thing could be said about World Chess Federation ranking system.

The H-rating is a tightly defined number that returns a player's win percentage in "two-player" parlance.  If you look at the FAQ, it is neither poorly defined or a pseudo-stat. On the contrary, I find it to be a rather elegantly derived number.

I like your Frankenstein analogy for the cumulative stat, because it is truly an attempt to sew together a bunch of disparate data.  There is no way to do it without weighting its components in arbitrary fashion.  It will be what it will be, and the top player will be whoever can best take advantage of how it's components are weighted, but we have control over that.