M57 wrote:

berickf wrote:

Amidon37 wrote:

OK, If it were really up to me I would do the following:

2) Board Ranking on each board would be determined by using option I from a previous discussion.

If you're suggesting we scrap GR entirely, I fear you'd be chasing a lot of people from feeling that they can get high rankings, myself included, because I am not a CP player!  A CP's only rank, to me, seems also to not reward newcomers, as to crack the ranks of the vets who had accumulated much of the CP over years of playing, would be a daunting task to say the least!  I think GR is really important to give newcomers to the site an immediate feeling that they can be competitive in a matter of months and not make them feel like ranking is largely a futile or extremely time consuming task of accruing enough CP to stack up against players who have been around for years already.  Plus, if as you say, CP is only given to the top ten players from each board, then how long will it take for a new player to see any score effect from their game play!

Note that Amidon is a proponent of Option I, which puts noobs right in the CP race as soon as they have a winning record.  I think there is a bit of misunderstanding that goes with the notion of dropping GR. The system would still exist - just not the "Global" part of it.  The system would be used on a board by board basis, and would be necessary to generate CPs.  We're just talking about throwing away a number that fluctuates all over the place.  Option I CPs are in a sense a much more accurate and stable aggregate.

I don't know what option I is, like I said, I stopped following the micro discussion regarding how GR and CP scores might be calculated, but, Amidon did say in his post that only the top ten players from each board would be given CP, which seems to differ from what you are saying here.  That said, however GR and/or CP were recalibrate to accrue could be applied throughout the greater framework that this thread is discussing, so, I have no problem with any readjustments down the road, just that that is not what this thread is about.

Quazimoto wrote:

As I see it. I like the idea of having 8 areas and I assume the Global score is a total ranking with all of them combined. At least that is how I would set it up, if it isn't. That is the score I would use for the top player. So you can have a top player in one area and not have a very high ranking in the Global area. As I mentioned before, the scoring system is not my thing.It appears you are trying to make it more fair to all the players.Is there a link I can read on how the scores are being measured? I would like to compare the GR, CP, and true systems and I would be able to tell you what I really think about each system. As in the first posting, I see no reason to continue in the direction the scoring system is heading into. M57 suggest that Cps are more accurate, what if you combine the two? How would that work. 

I also envision that once a balanced framework is constructed that it could be more easily combined into a fair aggregate score.  I think from the detour posts we have seen here thought that before an aggregate were attempted there seems to be a bigger issue on how GR and CP are calculated, which could then be applied throughout the framework I'm suggesting, to create an all encompassing, yet more fair and stable, system.  Say for instance the framework had all eight areas and then this option I was implemented to allow immediate accrual of CP's and then a trueskill was applied to give stable GR's.  At that point I think that people might be willing to discuss an aggregate?  Baby steps!

M57 wrote:

Quazimoto wrote:

 M57 suggest that Cps are more accurate, what if you combine the two? How would that work. 

I would clarify that "Option I" CPs are more accurate.

What would be the point of combining them when "Global" CP scores are already based on the individual board GR scores?  Currently, the "Global" GR score is just a meandering record of a player's ups and downs. It doesn't give us a sense of how diversified their play is, or depending on its range, even how good they are.

The GR system is flawed enough as is, and piling on just creates Frankenstats. Optimally a Trueskill-like GR system would give us a much better sense of a player's current strength on any given board.  This in turn should be used to generate Option I CPs, which should be "THE" statistic for assessing Global strength and achievement. 

I say the current GR system is flawed, but I believe it is less flawed at the single board level -  and as much as I would prefer to see a TrueSkill-like system replace GR at the board level - I can see that converting CPs to ICPs and trashing the GR would both strengthen and stream-line things, which is all the more important if the site is moving toward a multiple CP system across play-styles (Tournament, Public, Team, etc..).

Some will say GR are flawed while others, CP.  But, I would appreciate it if we could try and refrain from muddying up the thread by re-treading the same GR/CP discussions.  Those discussions have their place, and, I hope they can be applied to a more equitable framework at a later date, but, for now I would like to keep this discussion to the framework, what should be included/excluded from it.

I'm a proponent of the complete 8 rank framework, and, I even like to think that the heaviest criticizers of GR and/or CP might see that the 8 rank will only get stronger when and if GR and/or CP are amended down the road to address the weaknesses that they see in them.

I am not against CP, but, I am against a CP only system.  One reason is that many players like to come here to play one board and one board only, typically WGWF, and, with a CP only system you would kill their incentive to "rank" and in so doing, kill their investment in the site.  The only rank they'd be left to watch would be the WGWF board rank which is less visible and hard to crack on a board like that.  Instead of their global GR keeping them feeling competitive and invested, they'd more quickly grow bored and then stop coming around as often!  The ranking system might be largely flawed, but, it's biggest benefit is that it can keep people hooked if they see their screen name climbing a rank.  In that regard, the more the merrier!  If you can keep each and every player around and invested there is also more chance that they'll eventually try other things!

Thanks!

berickf wrote:

I don't know what option I is, like I said, I stopped following the micro discussion regarding how GR and CP scores might be calculated, but, Amidon did say in his post that only the top ten players from each board would be given CP, which seems to differ from what you are saying here.  That said, however GR and/or CP were recalibrate to accrue could be applied throughout the greater framework that this thread is discussing, so, I have no problem with any readjustments down the road, just that that is not what this thread is about.

-1 to a GR/CP all around system given the current underlying construct of the system.

The current GR underpinnings are inaccurate and the current CP system itself is too complex and top-weighted, so a multi-ranking system is just piling on, which just serves to magnify the weight of Cons as you have listed them.

My point is that the subject of this thread needs to be seen as being integrally tied to its scaffolding.

YuriZ wrote:

Ozyman wrote:

+1 to having a corresponding CP for each GR.  Makes sense to me.

+2!

Ozyman said it about as short and simple as it could be said +3!

I would like to add.  I used to like playing tournaments but since I came back to WarGear there are so few to choose from!  By having a CP rank for tournaments as well I was thinking that maybe that would encourage the creation of more tournaments?  As a standard member I can't create a tournament myself.  We standard players do rely on premium members to create tournaments for us to join.  Since if a player is paying for premium they might also be someone who's more interested in their CP, there is some sense to the assumption that a CP tournament rank might encourage the creation of more tournaments?  An alternative solution might be to open up tournament creation to standard players because as it is tournament selection is really poor these days.

bmasera wrote:

I would like to add.  I used to like playing tournaments but since I came back to WarGear there are so few to choose from!  By having a CP rank for tournaments as well I was thinking that maybe that would encourage the creation of more tournaments?  As a standard member I can't create a tournament myself.  We standard players do rely on premium members to create tournaments for us to join.  Since if a player is paying for premium they might also be someone who's more interested in their CP, there is some sense to the assumption that a CP tournament rank might encourage the creation of more tournaments? 

This is a strong additional argument for.  Mind you - I'm not really against it, I just believe it would be counterproductive in the short term ..and thus premature to institute it at this time. I'm concerned we're putting the cart before the horse.

M57 wrote:

berickf wrote:

I don't know what option I is, like I said, I stopped following the micro discussion regarding how GR and CP scores might be calculated, but, Amidon did say in his post that only the top ten players from each board would be given CP, which seems to differ from what you are saying here.  That said, however GR and/or CP were recalibrate to accrue could be applied throughout the greater framework that this thread is discussing, so, I have no problem with any readjustments down the road, just that that is not what this thread is about.

-1 to a GR/CP all around system given the current underlying construct of the system.

The current GR underpinnings are inaccurate and the current CP system itself is too complex and top-weighted, so a multi-ranking system is just piling on, which just serves to magnify the weight of Cons as you have listed them.

My point is that the subject of this thread needs to be seen as being integrally tied to its scaffolding.

I think that that's a flawed argument in that you are assuming that how gr and cp are calculated is the scaffolding.

Instead, I like to think that the framework of which ranks attract attention is the scaffolding in that it is these ranks which are followed that get players invested in the site. I then see the way GR and CP is calculated is simply the finishing to give the rank scaffolding a refined finish.

The multi GR/CP system does not "pile on" but instead is only meant to highlight areas of interest that players might want to compete over and should provide every player, old or new, somewhere to focus their efforts and be rewarded by rank.  Rank calculation refinement can come later, but, lets get the people in the door first and seeing reasons for sticking around!

Since I first stumbled upon WarGear it has seemed to me that the numbers of games and people have been a bit erratic but generally reducing, not growing.  The framework required to keeping players invested, therefore, and especially getting new players hooked and then invested, is first and foremost to me.  The finishing off of how gr and cp ranks are calculated is important, and will come.  But, if the player base is not growing and might in fact be diminishing, then does it even matter how GR or CP is calculated?  How much longer do we want "1st November 2011" to be the published peak of WarGear? If not, then how can we provide incentives to keep the old players around and to keep the new ones that stumble upon us playing here?

berickf wrote:

M57 wrote:

berickf wrote:

I don't know what option I is, like I said, I stopped following the micro discussion regarding how GR and CP scores might be calculated, but, Amidon did say in his post that only the top ten players from each board would be given CP, which seems to differ from what you are saying here.  That said, however GR and/or CP were recalibrate to accrue could be applied throughout the greater framework that this thread is discussing, so, I have no problem with any readjustments down the road, just that that is not what this thread is about.

-1 to a GR/CP all around system given the current underlying construct of the system.

The current GR underpinnings are inaccurate and the current CP system itself is too complex and top-weighted, so a multi-ranking system is just piling on, which just serves to magnify the weight of Cons as you have listed them.

My point is that the subject of this thread needs to be seen as being integrally tied to its scaffolding.

I think that that's a flawed argument in that you are assuming that how gr and cp are calculated is the scaffolding.

Instead, I like to think that the framework of which ranks attract attention is the scaffolding in that it is these ranks which are followed that get players invested in the site. I then see the way GR and CP is calculated is simply the finishing to give the rank scaffolding a refined finish.

The multi GR/CP system does not "pile on" but instead is only meant to highlight areas of interest that players might want to compete over and should provide every player, old or new, somewhere to focus their efforts and be rewarded by rank.  Rank calculation refinement can come later, but, lets get the people in the door first and seeing reasons for sticking around!

Since I first stumbled upon WarGear it has seemed to me that the numbers of games and people have been a bit erratic but generally reducing, not growing.  The framework required to keeping players invested, therefore, and especially getting new players hooked and then invested, is first and foremost to me.  The finishing off of how gr and cp ranks are calculated is important, and will come.  But, if the player base is not growing and might in fact be diminishing, then does it even matter how GR or CP is calculated?  How much longer do we want "1st November 2011" to be the published peak of WarGear? If not, then how can we provide incentives to keep the old players around and to keep the new ones that stumble upon us playing here?

If you are concerned about the players staying or leaving, put out a questionnaire to see why or what they want more of. If you go to the 8 section scoring and include all of the game types. This may give those who want to keep score that opportunity. Then with the global ranking and rolling all rankings by number of games or time, or both to keep people from sitting on the lead if they quit or put the game on hold. You might be surprised that the scoring system may not even matter to most players. As to which one you want to use, as long as it covers all types of games and game play, then that would be a step in the right direction. Tweaking the scoring system at a later date would also be a step in the right direction. If you go a little at a time then you can adjust what you want without completely changing the system. Just a thought.

Quazimoto wrote:

If you are concerned about the players staying or leaving, put out a questionnaire to see why or what they want more of. If you go to the 8 section scoring and include all of the game types. This may give those who want to keep score that opportunity. Then with the global ranking and rolling all rankings by number of games or time, or both to keep people from sitting on the lead if they quit or put the game on hold. You might be surprised that the scoring system may not even matter to most players. As to which one you want to use, as long as it covers all types of games and game play, then that would be a step in the right direction. Tweaking the scoring system at a later date would also be a step in the right direction. If you go a little at a time then you can adjust what you want without completely changing the system. Just a thought.

The problem with a questionnaire is you'd be asking players who are already here and would generally have their interests sunk into what they have already accomplished. That could be good for knowing how to keep players around, but, the bigger problem to me seems to be attracting fresh blood...  My strategy for that is the more hooks the better and hence this proposal! ;-) 

For the eight ranks, as well as an aggregate if it ever came to be, I like the idea of having a rolling average.  Good for keeping players from sitting on their leads and to keep them continually active on the site.  Big thumbs up on that!  The rolling average should be the "highlighted" rank in all areas that would be seen when looking up where you stand.  But, with a little digging, I'd still like to be able to view the "of all time" accumulated rank and perhaps a companion table listing the top twenty player's high tide mark of all time.  Sort of like a "hall of fame" for the non rolled score.  That could be where site leavers could still have a lasting presence and then people could still recognize their past glory when they come back on the scene!

Little by little a lot could be done.

I would like to thank everyone for the overwhelming positive response to this thread.  So far the biggest criticism has not even really been against the idea as much as it has criticized the underlying calculations for GR and/or CP and whether they believe a rank to be "garbage".  This criticism has led to the question of if we might be piling garbage on top of garbage by adding more ranks?  To that I can only say that these calculations can be revisited in the future, but, even as they are now, it has not stopped many players from still trying to rank high in these existing "garbage" ranks!  The other thing I might say is that while GR seemed to attract more criticism then CP, not a single additional GR rank is being proposed to be added to the ranking system!

I would like to address one thing that Amidon pointed out earlier, that he'd be more happy with the 8 rank proposal if the site had more players.  To that I'm thinking "chicken or egg?" and that we need to hold onto more new and old players in order to have more players.  I can't help thinking that focusing the framework towards a specialized CP score, even if it were "option I"ed and allowed a more immediate involvement, that it would still be a turn off especially to many new players who don't get how CP works, and result in not retaining new players as well as an eight rank system.  It also might discourage the existing GR players from sticking around.

In about two more weeks time after this thread has been able to run the course of the idea I would like to propose to Tom the concept of adding all eight ranks to the 5-rank general system that has recently been put up and has been linked to in this thread a few times already, to make it an 8-rank general system.  I think it would be a good step in the right direction to create a balanced ranking system framework that all the other micro proposals (like recalculating GR and/or CP, rolling averages for ranks), or macro proposals (like a weighted aggregate, HOF), can refine this framework in the future.  This thread has so far revealed no deleterious reasons against the idea of an 8-rank general system, so, I think that if Tom can put it up in preview it would help us all be able to see if we like it and also encourage new areas of competition!

As Tom likes consensus though, if one person who is adamantly against it says "no, I absolutely do not want to see a preview of the 8-rank general system", just like that, then I will take it as us not reaching that consensus and will not make any proposal to Tom.  In the mean time we should keep hashing through the concept!  All new discussions on the macro concept of rank framework are welcome here and if a better idea is formed by consensus then I'd gladly propose it instead!

Thank you everyone for your contributions and support of the idea!

Erick

berickf wrote:

I would like to address one thing that Amidon pointed out earlier, that he'd be more happy with the 8 rank proposal if the site had more players.  

I didn't mean to say that.  What I meant to say (I think) was that I think GR is a poor statistic for comparing players between boards.  (Obtaining a 1500 board score on some boards is just as hard as having a 1200 board score on other boards.)  Then the global GR is a bad statistic because it rewards certain boards/playing styles.

Even within a board the board score is suspect to a certain degree, but we don't have a large enough player base to really suss that out.  I would like to have rankings for each scenario, fog level, etc. on each board and some way of working them into the CP system.  This is not at all realistic at this point though.

Do I support 8 ranks?  eh..  I certainly support a separate CP system for team games.  I don't see the need to separate tournament games into their own separate categories, but I can go with it.  If I had my way I'd junk GR altogether, but if others want it I guess so.

I also very much support a "current players" ranking board of some sort.

Amidon37 wrote:

I also very much support a "current players" ranking board of some sort.

By this you mean a rolling average, or is this another idea all together?

I don't know enough about what a "rolling average" means to speak for or against it in particular.  I think more about "points in the last year" ranking.

I like the hall of fame idea and the 8 rank system. If it makes it a more even way to rate a player. The type of scoring system really is not the main issue if you are going to possibly change it. You need to have a good basic system or areas you want to rank and then if the present GR or CP system doesn't work efficiently enough, just change it to what you want to measure and how to measure it. Starting with a good solid base is the key to making this a success. As far as bringing new players in, you can give them a free month and allow them to play what they want, with the agreement they give their opinion of what is good or bad about the site. Just brainstorming. 

Amidon37 wrote:

I don't know enough about what a "rolling average" means to speak for or against it in particular.  I think more about "points in the last year" ranking.

So a one year rolling average then ;-)  Some people like to put it in terms of games and others in terms of time, but, a "points in the last year" would be a one year rolling average where every new day takes one day off the other end so the rank would only reflect the points gained/lost in the previous year.  I think you get it more then you realize!

>Some people like to put it in terms of games and others in terms of time

I think it needs to be in terms of time.  The point is to remove players who haven't played in a long time right? 

+1 to a one year moving average.