Coming out of the epic conversation of CP vs GR (Universal Aggregate) conversation it has been noted that the CP has areas that could be ironed out, to be more reflective of especially as it has shiny new rank badges to go with it.
I'll try to put the suggestions I've seen so far with some edits for clarity correct me if I miss quoted:
It's not:
+1 CP to every player over 1500 for every player over 1500 (just using the current border).
It would make everyone over the mark "in the points" and would make 1st place on wgwf worth around 120 CP's to first place.
Or, + X% to each player for each player over 1500. (No idea the number)
Extend the current threshold: to say 2000 and pay out the top 20?
Ozy:
Since WGWF is really the special case driving all this, I suggest we treat it as a special case (all #s pulled from my ass):
#1) Increase CP scale for all boards to give CP to the to 20. Start at +30 for 1st place and 2000 points, and go down from there.
#2) Give WGWF it's own special CP scale. Make 1st place @ 3000 pts worth 100, and go down from there, giving points to the top 50.
Ratsy:
+1 to this point. I also really harmonized with the idea of finding a way to award more CP for the more played maps, and less for the less played. Bottom line: more competition - greater champion
M57:
+1 Usage Weighted CPs - with or without WGWF being a special case. My inclination would be to have a some kind of regressive system such that the uber-popular boards are not too influential.
E.g., give the top 16 boards extra CPs linearly:
WGWF top player = 100
Colossal Crusade top player = 95
..90, 85, 80,.. ..30, 25, 20, 20, 20,..
Hugh:
WGWF is worth more than an obscure map with no competition, but Invention is worth more than that, because it's actually hard AND competitive. (I'm not making a proposal here - just stating an ideal that I have in my mind.)
Right now, this is a subjective opinion of mine. I don't just mean mechanics. Once past the noob level, I believe it requires greater skill to play well than WGWF. It at least seems that way, but it is just a subjective judgement.
On obtaining objectivity - This would require research, but one of the things I read about the Trueskill ranking algorithm is that it can quantify the "levels of skill" a game has. If there is any calculation that can do that, I know a lot of us would find that sort of stat very interesting!!
After pondering some of the options I hear I'm leaning towards keeping the current system, but extending the current threshold: to say 2000 and pay out the top 20 with a top score of 40? 50?
The system is already in place and would just need a slight mod to the numbers.
This help address the issue of more competitive boards (WGWF, Colossal, Civil War), being undervalued.
And, would reward a bit better some of those one board dynamos that are so dominating a board they are up over 2000 when everyone else hasn't broke 1500.
It doesn't totally address the special case of WGWF, but it would help. I also liked the idea someone said of giving that board a gold plated badge. Or maybe two. One for the current champion that disappears and a permanent one for any previous champion.
I think your final proposal is reasonable, and most importantly relatively simple. Thanks for making this thread itsnotatumor. I was contemplating doing the same last night. That other thread is such a morass, and this seemed like a relatively easier subject to reach consensus on..
Just to tweak it a bit, I'd say go all the way to 2500 & pay out the top 25 starting at 50 points. This will help make boards that are played a lot worth more. On a board that isn't played much it will be very hard (nigh impossible) to reach 2500, so that will mean players will get less points for top positions in less played boards than in the most popular ones.
Right now WGWF is the only board with a top player above 2500, so it would be the only board where #1 would earn the full 50 points. Colossal crusade is close behind. Other boards will earn a lot less depending on how the scale continues into the lower levels.
For comparison here is the current CP calculation:
I suggest something like:
It seems like these #s were pretty arbitrary to begin with, and as the site has grown, they probably need to grow too.
So for the sake of complete thinking it through:
How many points will the top scorers have after this system is implemented?
Does it make it more or less easy to "game the system"?
Does it change the resolution of the cp system for the better or worse? (in the sense that if got 10 more points right now, current system, I'd gain 5 ranks, which I could do with a couple games on one not-so-played-board like escherdiles)
Does that make CP more or less accessible to everyone - including non-premium, or not so active players, or those that like to specialize?
How much harder does it make it to climb the CP ladder? i.e: Do I now have to play and win x number of games of WGWF in order to be top 10 CP, or do I just have to beat the good players?
Will it put more players on the CP ladder?
Will the system still encourage playing lots of varieties of boards, as it is currently the only system in place to do so?
Is this somehow capturing Hughs thoughts about boards that are hard and competitive?
Just to put some context to Ozy's proposed new sliding CP scale, I calculated the CP totals for Cona (the current CP leader) and Toto (currently #1 on WGWF).
Cona's total would change from 507 to 482. He would pick up some CPs on boards that he is ranked above 1600 and a couple where he is out of the top 10, but he gets dinged a lot on boards where he is #1 with 1500.
Toto's total would change from 102 to 237. He gets the 50 from WGWF and picks up 36 from Antastic. He also picks up points from other world maps that are played a lot.
Thanks for the #s CK. IMO those results still seem pretty reasonable, and hopefully address some of the complaints about WGWF not getting enough weight.
Good questions ratsy. CK answers the first one, and I'll try on some of the others.
>Does it make it more or less easy to "game the system"?
I would say harder, because you have to play more games to earn more points. It's still just as easy to pick up just a point or two on boards that aren't played much, but that is offset by players getting a lot more points on boards that are played a lot. Previously getting to 1500 on a board got you 20, now just 15, so getting those 20 points on a less played board is going to be a bit harder now.
>Does it change the resolution of the cp system for the better or worse? (in the sense that if got 10 more points right now, current system, I'd gain 5 ranks, which I could do with a couple games on one not-so-played-board like escherdiles)
It will lead to a lot more CP points overall, so any individual CP would be worth less. So in that sense getting a few points here and there on less popular boards will not be enough to get you up to the level of the leaders on a popular board like WGWF or antastic.
>Does that make CP more or less accessible to everyone - including non-premium, or not so active players, or those that like to specialize?
Not sure about this one. I'm not sure if it helps with non-premium members, although based on berickf's analysis, I guess it would since the non-premiums usually play more on the popular boards, so more chance for them to get CP.
>How much harder does it make it to climb the CP ladder? i.e: Do I now have to play and win x number of games of WGWF in order to be top 10 CP, or do I just have to beat the good players?
Again not sure... I think how long it takes to climb the CP ladder should not necessarily be a factor. I'm fine with it taking a while for someone who just joined to climb the ranks to compete with someone who has playing for 2 years.
>Will it put more players on the CP ladder?
Yes, because there will be more points. In particular a board where the CPs are maxed out (like WGWF) would contribute 394 CP, while a typical less played board where the max score is 1500 would only contribute up to 70 CP. (Max 81 CP per board previously, so WGWF would now be worth almost 5x what it was previously)
>Will the system still encourage playing lots of varieties of boards, as it is currently the only system in place to do so?
I think it will. A good player can still try to pick up lots of points across a variety of boards, and in fact for someone like Cona who had maxed out a lot of boards at 1500, it would encourage them to go back to those boards and keep reaching.
>Is this somehow capturing Hughs thoughts about boards that are hard and competitive?
No. I don't think so. If you did want to try and weight harder boards more uou could have a multiplier for whether the board is green/yellow/red. For example green boards earn CP based on the table above. Yellow boards get those CP * 1.33, Red boards get those CP * 1.66. The additional reach of the CP takes care of the competitive part of things, the multiplier takes care of the 'hard'. This is all so subjective, I'm ambivalent about this idea, but I think it would work ok.
>Right now WGWF is the only board with a top player above 2500, so it would be the only board where #1 would earn the full 50 points.
FYI - I guess this is wrong. Antastic would also be worth the full amount, maybe there are others.
Ozyman wrote:For comparison here is the current CP calculation:
- 1500+ score - 20 Championship Points
- 1450+ score - 15 Championship Points
- 1400+ score - 12 Championship Points
- 1350+ score - 10 Championship Points
- 1300+ score - 8 Championship Points
- 1250+ score - 6 Championship Points
- 1200+ score - 4 Championship Points
- 1150+ score - 3 Championship Points
- 1100+ score - 2 Championship Points
- 1050+ score - 1 Championship Points
I suggest something like:
- 2500+ score - 50 Championship Points
- 2400+ score - 45 Championship Points
- 2300+ score - 40 Championship Points
- 2200+ score - 36 Championship Points
- 2100+ score - 33 Championship Points
- 2000+ score - 30 Championship Points
- 1900+ score - 27 Championship Points
- 1800+ score - 24 Championship Points
- 1700+ score - 21 Championship Points
- 1600+ score - 18 Championship Points
- 1500+ score - 15 Championship Points
- 1450+ score - 13 Championship Points
- 1400+ score - 11 Championship Points
- 1350+ score - 9 Championship Points
- 1300+ score - 7 Championship Points
- 1250+ score - 5 Championship Points
- 1200+ score - 4 Championship Points
- 1150+ score - 3 Championship Points
- 1100+ score - 2 Championship Points
- 1050+ score - 1 Championship Points
It seems like these #s were pretty arbitrary to begin with, and as the site has grown, they probably need to grow too.
+1 Looks good to me.
Ozyman wrote:>Right now WGWF is the only board with a top player above 2500, so it would be the only board where #1 would earn the full 50 points.
FYI - I guess this is wrong. Antastic would also be worth the full amount, maybe there are others.
Civil War, World War, and Capitalism are over 2500 as well. Though with Cap only Recon will get the bump (arguably deserved).
Both Colossal and A&A are only 40 points away.
Though for A&A there are some big point gaps after the lead positions.
Hmmmm, I'm not sure I agree with all this as I think it will grow the gap between players rather than close it or truly better define it. I think the reason that a lot of the current CP chasers don't have a score much above 1500 is due to the fact that there is nothing great to obtain (unless they have to overtake a spot), not because they are not willing to grind it out to 2000-2500+. So I could see this as giving players a higher goal to shoot for (which is a good thing), but giving the top players a chance to get 30+ more medals will grow the gap to both take out a top player on a board and also the then cumulative CP counts. I'd much rather have a goal of hitting 1550 to grab 20 CPs and drop someone in the Rankings down rather than trying to hit 2500+ to grab 50 and drop the person. I could see this as forcing even more ("bad/unpopular") boards to be played for high score counts, which is not a bad thing either but was brought up as a potential negative to CP grabbers play.
Soooo, what I would look at proposing is possibly a "bonus" system at one or more levels in which those CPs are guaranteed, which would be in addition to the current system. So anyone that hits 1750+ gets 5 CPs (while maintaining 1750+), 2000+ gets 8, 2500+ gets 12 (I just pulled those numbers out of the air) in addition to the 20 CPs for 1500 and #1. I see this as promoting some set goals with a guaranteed payout along with going beyond the current 1500 mark.
Is this site equivalent to a military dictatorship or is it a democracy? If it is democratic then the privates are the clear cut majority by far and should have the final say. I only see "CP" Generals doing all the posting and decision making keeping their flawed coveted "Championship Points" in mind. Everyone needs to think out of the box and come up with a totally new ranking system to either replace what's in place or to be kept as a totally separate ranking that can be compared to rankings that are currently in place.
uhh, its truly a dictatorship. Commander in Chief has the final say. End of discussion.
We just like to dream up the things that we like and we hope in the end that Tom likes them too and puts them into action (he's good like that too). Anyone and everyone can weigh in here. It's pretty democratic.
You see alot of Generals putting their two cents in because they spend alot of time here, and care about what happens to it and making it better. That's all.
Ozyman wrote:For comparison here is the current CP calculation:
- 1500+ score - 20 Championship Points
- 1450+ score - 15 Championship Points
- 1400+ score - 12 Championship Points
- 1350+ score - 10 Championship Points
- 1300+ score - 8 Championship Points
- 1250+ score - 6 Championship Points
- 1200+ score - 4 Championship Points
- 1150+ score - 3 Championship Points
- 1100+ score - 2 Championship Points
- 1050+ score - 1 Championship Points
I suggest something like:
- 2500+ score - 50 Championship Points
- 2400+ score - 45 Championship Points
- 2300+ score - 40 Championship Points
- 2200+ score - 36 Championship Points
- 2100+ score - 33 Championship Points
- 2000+ score - 30 Championship Points
- 1900+ score - 27 Championship Points
- 1800+ score - 24 Championship Points
- 1700+ score - 21 Championship Points
- 1600+ score - 18 Championship Points
- 1500+ score - 15 Championship Points
- 1450+ score - 13 Championship Points
- 1400+ score - 11 Championship Points
- 1350+ score - 9 Championship Points
- 1300+ score - 7 Championship Points
- 1250+ score - 5 Championship Points
- 1200+ score - 4 Championship Points
- 1150+ score - 3 Championship Points
- 1100+ score - 2 Championship Points
- 1050+ score - 1 Championship Points
It seems like these #s were pretty arbitrary to begin with, and as the site has grown, they probably need to grow too.
I've been following this discussion off and on so bear with me...I don't think the above is going to really help. It will just buy some time before we have the exact same discussion we're having now. By extending the CP max it will just take awhile longer for the top ranked players to get to the new max threshold.
This may have been proposed before but why not have a simple calculation that weights both difficulty and popularity of the boards played with the current CP range? The difficulty factor would be a tiered system based on the Red, Green, Yellow lights and popularity would be based on total all time # of plays you can see in the charts. Popularity could be normalized/fractionalized to account for really popular boards (or not normalized/fractionalized since we're saying that those are the most competitive) and to avoid skewing the results.
For example:
Green difficulty = x1
Yellow difficulty = x2
Red difficulty = x3
WGWF multiplier: x.9
CC: x.1
AxA: x.05
The total CP would then be: CP x Difficulty factor x Popularity factor. I'm just making these numbers up but you get the idea.
The advantages to this system is:
Though it sounds a bit too convoluted for most people (the rank and file) to take the time to understand, I like the general idea of ATH's proposal, but I still have a big problem with how red yellow green hard/easy can be determined.
For example, Hex may be the simplest board to play on this site.. you don't have to worry about dice odds - you can only play one piece at a time - there's no movement, fortifying, cards, and all you have to do to win is make a line. Yet, I doubt there are many who would say the board is "easy" to win at.
There have been discussions about determining criteria for 'difficulty.' and these should probably be revisited.
Difficulty is a loaded term. Hex could be considered difficult simply because it is different. For example, we could say that any/every change in the standard WGWF rules could be considered confusing and therefore increase a board's difficulty rating.
M57 wrote:but I still have a big problem with how red yellow green hard/easy can be determined.
This is my concern as well with using the "Difficulty" information on a Board. Currently I believe it is only myself (for better or worse) that sets this setting. I do try and keep like Boards the same and I try and think through the "difficulty" of a Board, but I'm sure there is subjectivity in the setting (although I'm always up for looking at changing/updating any Board).
Also, the current Difficulty is more of a judge of playing the Board and knowing it rather than how difficult it is to move up the Rankings, so we would probably really need some sort of different "Difficulty" field.
Yes, Difficulty of a board should be different from Competition/Popularity. However, I think even without a new ranking system the Difficulty judgement should be more rigorous and systematic. There should be some incentive for each player to have input into the Difficulty rating of a board, since it's there to help them in the first place.
So, assuming we would do that anyways, it makes sense to use this for CP rating.
Yertle wrote:M57 wrote:but I still have a big problem with how red yellow green hard/easy can be determined.
This is my concern as well with using the "Difficulty" information on a Board. Currently I believe it is only myself (for better or worse) that sets this setting. I do try and keep like Boards the same and I try and think through the "difficulty" of a Board, but I'm sure there is subjectivity in the setting (although I'm always up for looking at changing/updating any Board).
Also, the current Difficulty is more of a judge of playing the Board and knowing it rather than how difficult it is to move up the Rankings, so we would probably really need some sort of different "Difficulty" field.
Here's a related older discussion on the Complexity/Difficulty issue for anyone who's interested.
Babbalouie wrote:Is this site equivalent to a military dictatorship or is it a democracy? If it is democratic then the privates are the clear cut majority by far and should have the final say. I only see "CP" Generals doing all the posting and decision making keeping their flawed coveted "Championship Points" in mind. Everyone needs to think out of the box and come up with a totally new ranking system to either replace what's in place or to be kept as a totally separate ranking that can be compared to rankings that are currently in place.
Usually the way it works is that we all discuss in the forums. Tom is either ignoring us or following along, but it's hard to tell. If we come to some sort of consensus and most importantly if Tom likes the idea and has the time to make the changes, then he modifies/upgrades things.
If we can get to the point where most of the people agree, someone can PM Tom, and usually he'll go along with our recommendations when he gets the time.
You also have to realize that their is an outstanding wish list a mile long of changes/improvements we'd like, but Tom is just one (very busy) guy, so even if we come up with a consensus idea, and he likes it, it might be 6 months before it gets changed.
So mostly we're discussing this for the intellectual exercise and fun of discussion. That's one reason it doesn't pay to get to emotionally involved. If wargear ends up better because of our discussion that's a bonus, but I wouldn't assume their is necessarily going to be any changes to the way CP (or an aggregate) is calculated, and if we all can't come to some agreement, chances are even less.
AttilaTheHun wrote:Ozyman wrote:For comparison here is the current CP calculation:
- 1500+ score - 20 Championship Points
- 1450+ score - 15 Championship Points
- 1400+ score - 12 Championship Points
- 1350+ score - 10 Championship Points
- 1300+ score - 8 Championship Points
- 1250+ score - 6 Championship Points
- 1200+ score - 4 Championship Points
- 1150+ score - 3 Championship Points
- 1100+ score - 2 Championship Points
- 1050+ score - 1 Championship Points
I suggest something like:
- 2500+ score - 50 Championship Points
- 2400+ score - 45 Championship Points
- 2300+ score - 40 Championship Points
- 2200+ score - 36 Championship Points
- 2100+ score - 33 Championship Points
- 2000+ score - 30 Championship Points
- 1900+ score - 27 Championship Points
- 1800+ score - 24 Championship Points
- 1700+ score - 21 Championship Points
- 1600+ score - 18 Championship Points
- 1500+ score - 15 Championship Points
- 1450+ score - 13 Championship Points
- 1400+ score - 11 Championship Points
- 1350+ score - 9 Championship Points
- 1300+ score - 7 Championship Points
- 1250+ score - 5 Championship Points
- 1200+ score - 4 Championship Points
- 1150+ score - 3 Championship Points
- 1100+ score - 2 Championship Points
- 1050+ score - 1 Championship Points
It seems like these #s were pretty arbitrary to begin with, and as the site has grown, they probably need to grow too.
I've been following this discussion off and on so bear with me...I don't think the above is going to really help. It will just buy some time before we have the exact same discussion we're having now. By extending the CP max it will just take awhile longer for the top ranked players to get to the new max threshold.
This may have been proposed before but why not have a simple calculation that weights both difficulty and popularity of the boards played with the current CP range? The difficulty factor would be a tiered system based on the Red, Green, Yellow lights and popularity would be based on total all time # of plays you can see in the charts. Popularity could be normalized/fractionalized to account for really popular boards (or not normalized/fractionalized since we're saying that those are the most competitive) and to avoid skewing the results.
For example:
Green difficulty = x1
Yellow difficulty = x2
Red difficulty = x3
WGWF multiplier: x.9
CC: x.1
AxA: x.05
The total CP would then be: CP x Difficulty factor x Popularity factor. I'm just making these numbers up but you get the idea.
The advantages to this system is:
- Doesn't require a special case for any board
- Encourage rigor and accuracy in board difficulty ratings, which should be done anyways.
- Relatively simple to understand
- Doesn't just delay the inevitable recurrence of this discussion (maybe too optimistic I know!)
>don't think the above is going to really help. It will just buy some time before we have the exact same discussion we're having now.
I basically agree, but I think it will buy a lot of time. It's taken several years to get to this point, it will probably take several more for the above chart to get out of date. And so what, then we update it again.
>By extending the CP max it will just take awhile longer for the top ranked players to get to the new max threshold.
It's not just about the top ranked players. It's about the guy who is ranked 12th on WGWF and is getting nothing for it.
--------
That said, I think some automatically scaling system is fine too, if we can come up with the math.