A quick thought regarding the semantics of the term Championship Points, which has us dealing with a somewhat conflicted and polarized continuum of understanding/goals. Here are the two camps of thought as I see them:
Clearly, I'm a proponent of the latter. As it is, the top 3 players on a given board get their avatars posted on the board description page, and the top board spots also get acknowledged on each of the players' home pages respectively. That should be enough. I want to know who can play competitively across the site more than I want to know who specializes the most. I see no reason to overly compensate those who go out of their way to "maintain position" in the pursuit of points. The position itself should be reward enough.
And obviously I was suggesting the former. I think if the top 3 spots are compensated more (and the top spot even more), then it gives more reason to fight for the top. Otherwise, the top several players on a board could have the same or close to the same CP from that board. Why fight for #2 when #3 gives you the same CP?
I think a more progressive system than we currently have sounds great, but why go to the opposite extreme?
Korrun wrote:And obviously I was suggesting the former. I think if the top 3 spots are compensated more (and the top spot even more), then it gives more reason to fight for the top. Otherwise, the top several players on a board could have the same or close to the same CP from that board. Why fight for #2 when #3 gives you the same CP?
I think a more progressive system than we currently have sounds great, but why go to the opposite extreme?
Like I said, the top players on each board get their kudos in other ways. We could have medals and awards, and maybe even a separate system for keeping track of the "position game," but when it comes to Rank (General, Colonel, etc.) I maintain that a pure egalitarian system is much more equitable, enabling pure comparison, especially among medium and lower echelon players.
Anyway, the obvious compromise is the one you recently suggested. Pile on by giving the top spots additional points. The problem is how do you do that in a way that is anything but arbitrary? Answer the questions I posed on the ICPs thread regarding which spots, how much, etc.. It's a political quagmire.. Who's going to decide? If I'm in the top dozen on a bunch of boards, I'm going to lobby for the top 12 getting extra points.. If I'm in the top 10% of all the boards on which I'm competitive, I'm going to lobby for that.. Why don't we just give the very top spot extra points? ..or give 'em all to Mad Bomber and be over it..(for the record - that was a joke ;) After all, there can only be one champion. I just don't see the point. What other sports or games use inflated non-linear systems to determine their top players?
I tend to play my own boards more or less exclusively, which should put me in the specialist category. I haven't checked against my peers but I'm going to guess that my CP rank will fall in a pure ICP system, and I'm OK with that.
M57 wrote:Korrun wrote:And obviously I was suggesting the former. I think if the top 3 spots are compensated more (and the top spot even more), then it gives more reason to fight for the top. Otherwise, the top several players on a board could have the same or close to the same CP from that board. Why fight for #2 when #3 gives you the same CP?
I think a more progressive system than we currently have sounds great, but why go to the opposite extreme?
Like I said, the top players on each board get their kudos in other ways. We could have medals and awards, and maybe even a separate system for keeping track of the "position game," but when it comes to Rank (General, Colonel, etc.) I maintain that a pure egalitarian system is much more equitable, enabling pure comparison, especially among medium and lower echelon players.
Anyway, the obvious compromise is the one you recently suggested. Pile on by giving the top spots additional points. The problem is how do you do that in a way that is anything but arbitrary? Answer the questions I posed on the ICPs thread regarding which spots, how much, etc.. It's a political quagmire.. Who's going to decide? If I'm in the top dozen on a bunch of boards, I'm going to lobby for the top 12 getting extra points.. If I'm in the top 10% of all the boards on which I'm competitive, I'm going to lobby for that.. Why don't we just give the very top spot extra points? ..or give 'em all to Mad Bomber and be over it..(for the record - that was a joke ;) After all, there can only be one champion. I just don't see the point. What other sports or games use inflated non-linear systems to determine their top players?
I tend to play my own boards more or less exclusively, which should put me in the specialist category. I haven't checked against my peers but I'm going to guess that my CP rank will fall in a pure ICP system, and I'm OK with that.
If we adopt the multiple scoring method that was implemented in beta a while back, then General, Colonel, etc. ranks won't be based on just CP. I fully support that system. It would also make this discussion on equity slightly less important.
Anyway the two reasons I have are: 1. making playing new boards worth more. In your proposal, there would be very little benefit score wise to playing new boards or unpopular boards. 2. Making the top spots more contentious.
For reference this is the multiple ranking system: http://www.wargear.net/rankings/show4ranks and the discussion about it: http://www.wargear.net/forum/showthread/3817p5/Changing_CP_Consensus
I think it just stalled out from needing artwork done.
Korrun wrote:Anyway the two reasons I have are: 1. making playing new boards worth more. In your proposal, there would be very little benefit score wise to playing new boards or unpopular boards. 2. Making the top spots more contentious.
How do Option I points make the top spots contentious?? It is your add-on suggestion that weights the top spots on each board, making them more 'sought after'.
As for how playing new boards is concerned, I'd have to do a more thourough analysis, but my quick example of the top three current players strongly suggests that those who diversify their play do much better in a CPI system.
Maybe I don't understand your point. The problems you state above are exactly what Option I addresses.
M57 wrote:Korrun wrote:Anyway the two reasons I have are: 1. making playing new boards worth more. In your proposal, there would be very little benefit score wise to playing new boards or unpopular boards. 2. Making the top spots more contentious.
How do Option I points make the top spots contentious?? It is your add-on suggestion that weights the top spots on each board, making them more 'sought after'.
I think you misunderstood? I think the top spots should be more 'sought after'. My suggestion would help that.
As for how playing new boards is concerned, I'd have to do a more thourough analysis, but my quick example of the top three current players strongly suggests that those who diversify their play do much better in a CPI system.
You are absolutely right. Apparently my actual opinion is almost the opposite of what I thought it was. What I was trying to say is that I would prefer to see a #1 player on a less popular board get more CP than a middle level player on a moderately popular board. But that is more specializing and less diversity (but not too much less diversity).
Maybe I don't understand your point. The problems you state above are exactly what Option I addresses.
Option I does help with the diversity point, but not the contention for top spots. You could have a pretty good CP score without ever being in the top 3 on a board. To increase CP with Option I your best bet would be to play every single board up to or slightly past your equilibrium.
One last thought:
With your current Option I proposal (Board GR - 1000)/20:
The #1 spot on Battle of Waterloo would be worth 15 CP. #298 through #315 on WGWF would also get 15 CP. I don't think that is equitable. For WGWF player #301 won 5 and lost 73. #304 won 3 and lost 0. Does that sound similar to the top 3 players for Battle of Waterloo?
The shit emanated later when MB got on my back with some biased and untrue accusations with regard to dueling that is clearly not supported by the data.
Stop trying to pretend your shit don't stink. Stop trying to spin yourself as the victim. Your post #21 was a wordy way of saying High GR players good, High CP players sucky system gamers. MB then responded in kind though more succinctly.
Kinda reminded me of this:
https://www.youtube.com/watch?v=N_GjFMzVT7c
Just because you use lots of words when you're trash talking doesn't mean we don't notice.
Just like you use use lots of big words to show that you're right and other people are wrong whether the totality of data actually supports you or not. EVERY rating is gameable.
But, this thread is not about who is better, or which rating is better, or getting rid of CP, but how to make CP more equitable, less arbitrary, while still promoting diversity and innoation in boards.
Korrun wrote:One last thought:
With your current Option I proposal (Board GR - 1000)/20:
The #1 spot on Battle of Waterloo would be worth 15 CP. #298 through #315 on WGWF would also get 15 CP. I don't think that is equitable. For WGWF player #301 won 5 and lost 73. #304 won 3 and lost 0. Does that sound similar to the top 3 players for Battle of Waterloo?
One issue Option I does not address well is the discrepancy that happens with winning 2 player games vs games with many more players. Still the best way to earn points will be in playing 6-10 player games. I think only a trueskill-like system could address that. (at least of what has been talked about.) But we do have some players who have earned a high GR by focusing on 2-player boards so that could still happen here.
I have some sympathy for your issue of awarding extra credit for being in the top ranks, but I'd like to see it in action first without such add-ons to see how it looks.
Korrun wrote:Option I does help with the diversity point, but not the contention for top spots. You could have a pretty good CP score without ever being in the top 3 on a board.
It's unlikely that someone would be #3 on a ten boards and not #1 on any, but regardless, what's wrong with that?
To increase CP with Option I your best bet would be to play every single board up to or slightly past your equilibrium.
Exactly! Exactly what we want. Are you worried that a player might just stop playing because they've hit equilibrium?
One last thought:
With your current Option I proposal (Board GR - 1000)/20:
The #1 spot on Battle of Waterloo would be worth 15 CP. #298 through #315 on WGWF would also get 15 CP. I don't think that is equitable.
I do. What would be more equitable in your opinion? I'm going to speculate that you think it's easier to get to the #315 spot on WGWF. While technically it may be harder to get to the #1 spot on BoW, I would submit that it's just as easy if not easier to get to 1318 (currently #1) on Waterloo, than it is to get to 1318 on WGWF. (Assume the boards are equally difficult and equally fair - which we know isn't really the case because BoW is a dueling map).
For WGWF player #301 won 5 and lost 73. #304 won 3 and lost 0. Does that sound similar to the top 3 players for Battle of Waterloo?
Not really a fair comparison.. Like I said. BoW is a dueling map.. You can't rise as 'quickly' because you can't play in a game with 16 players - not to mention the board just doesn't get played that much and a number of other map-specific reasons.
M57 wrote:Korrun wrote:Option I does help with the diversity point, but not the contention for top spots. You could have a pretty good CP score without ever being in the top 3 on a board.
It's unlikely that someone would be #3 on a ten boards and not #1 on any, but regardless, what's wrong with that?
To increase CP with Option I your best bet would be to play every single board up to or slightly past your equilibrium.
Exactly! Exactly what we want. Are you worried that a player might just stop playing because they've hit equilibrium?
Not particularly. Just thinking out loud. Overall I am liking Option I more the more I think about it. Still would like a extra bonus for top spots though. In the divide by 20 system (15 CPs to 100 CPs for #1 positions) I would say +20 for #1, +10 for #2, +5 for #3.
One last thought:
With your current Option I proposal (Board GR - 1000)/20:
The #1 spot on Battle of Waterloo would be worth 15 CP. #298 through #315 on WGWF would also get 15 CP. I don't think that is equitable.
I do. What would be more equitable in your opinion? I'm going to speculate that you think it's easier to get to the #315 spot on WGWF. While technically it may be harder to get to the #1 spot on BoW, I would submit that it's just as easy if not easier to get to 1318 (currently #1) on Waterloo, than it is to get to 1318 on WGWF. (Assume the boards are equally difficult and equally fair - which we know isn't really the case because BoW is a dueling map).
Okay I'll race you. You get to 1318 on BoW (which shouldn't be too hard since it is your board) and I'll try to get to 1318 on WGWF first.
I'd hate to see the 1 point per net game paradigm get mauled by arbitrary add-ons.. So, 10 points for #1, huh? Regardless of whether you have 1100 or 3000 points to get there? I assume you also want to have minimums. Effectively, what you're talking about is adding Option I to the existing system.
M57 wrote:I'd hate to see the 1 point per net game paradigm get mauled by arbitrary add-ons.. So, 10 points for #1, huh? Regardless of whether you have 1100 or 3000 points to get there? I assume you also want to have minimums. Effectively, what you're talking about is adding Option I to the existing system.
FWIW, I'd rather add I-points to the existing CP system (or some version of the existing alla korrun's suggestion) rather than leave it alone. Even though this would make things even more convoluted, at least it adds an egalitarian element. In the sense that the 1 point per game ethos would become muddled, it doesn't matter as much what the multiplier/divisor becomes. A more important consideration would be balancing the two components.
Can folks perhaps weigh in on this thread?
Firstly, would you endorse ICPs at all? ..and if so, would you prefer them straight up alone (1 point per 20 GR), or would you like them 'enhanced' with something along the lines of korrun's suggestions?
M57 wrote:M57 wrote:I'd hate to see the 1 point per net game paradigm get mauled by arbitrary add-ons.. So, 10 points for #1, huh? Regardless of whether you have 1100 or 3000 points to get there? I assume you also want to have minimums. Effectively, what you're talking about is adding Option I to the existing system.
FWIW, I'd rather add I-points to the existing CP system (or some version of the existing alla korrun's suggestion) rather than leave it alone. Even though this would make things even more convoluted, at least it adds an egalitarian element. In the sense that the 1 point per game ethos would become muddled, it doesn't matter as much what the multiplier/divisor becomes. A more important consideration would be balancing the two components.
Can folks perhaps weigh in on this thread?
Firstly, would you endorse ICPs at all? ..and if so, would you prefer them straight up alone (1 point per 20 GR), or would you like them 'enhanced' with something along the lines of korrun's suggestions?
To start, I really like ICP's. Seems the most fair of the easily attainable ranking systems...so +1 for ICP's in general.
As far as Korrun's suggestion, I like his idea of awarding extra points to the top spots as it does drive more competition for a rise to the top. What I don't like is the flat amount of points awarded (+20, +10, +5), as the inequity of that is one of the reasons we're trying to implement a new system in the first place. Instead, how about a percentage of that player's ICP's awarded for the top spots? Something like +3% for #1, +2%, for #2, +1% for #3...or +6%, +4%, +2%. I don't really have the time now to figure out how the math would work out on that, but it would solve the inequity of awarding the same amount of points for the hard work to get to 3000 on WGWF to the comparatively easier work of reaching the top spot on a less popular board.
itsnotatumor wrote:The shit emanated later when MB got on my back with some biased and untrue accusations with regard to dueling that is clearly not supported by the data.
Stop trying to pretend your shit don't stink. Stop trying to spin yourself as the victim. Your post #21 was a wordy way of saying High GR players good, High CP players sucky system gamers. MB then responded in kind though more succinctly.
Kinda reminded me of this:
https://www.youtube.com/watch?v=N_GjFMzVT7c
Just because you use lots of words when you're trash talking doesn't mean we don't notice.
Just like you use use lots of big words to show that you're right and other people are wrong whether the totality of data actually supports you or not. EVERY rating is gameable.
But, this thread is not about who is better, or which rating is better, or getting rid of CP, but how to make CP more equitable, less arbitrary, while still promoting diversity and innoation in boards.
+1
Completely agree. This is exactly what I was saying. Every stat is gameable! I even said that in my post and did so expressively for how GR is gameable when I wrote: "how disciplined they are to sticking to putting them self in as many optimum game situations as possible". We all know how CP is gameable hence the point of this discussion and why it deviates even more greatly from h-rank. h-rank is the least gameable and hence why I was using it to discern the 'skill' base of the ranking systems. h-rank does concede that high GR players do exhibit more skill in attaining their positions, but, that was not meant as "trash talk", just a fact.
My question stemming from this, however, is simply to ask how CP can have more of a relationship to skill then it does now, not that GR is better, as you seem to have taken from my point. M57 pointed me in the direction of option I, now called CPI, which does seem to maybe have some teeth to streamlining this by rewarding more CP the higher a board GR gets and thus stressing skill to continue to push one's CP up once a player has gobbled up all the low lying fruit. So, indeed, it seems like the egalitarian CPI version with/without a low percentage perk for the top x positions might indeed push CPI to being more skill based and allow skill players to continue to accrue CPI on boards by pushing their board GR to where other less skilled players are unable to achieve.
Of course the broad base of CPI would still depend on playing many boards and thus put pressure against specialists, but, on a board by board basis it actually should reward skill more and thus addresses exactly what I was asking if it was possible to address when I wrote my now infamous post #21. Anyway, I'm really just trying to contribute to the conversation and since many of the points I brought up are being addressed by those still on topic, I am happy that I brought those points to the surface.
And yes, my shit does stink...
blah blah go for a ride blah blah outside blah blah
that's about how all these CP/GR/Option-whatever threads read to me. It is wanting much more attention than I really want to put into it.
But what I have gleaned from it is ICPs are better than what there is now, and that if we go that way I don't want some arbitrary number added to the top whatever spot(s). Go straight up. IF we find out later that we need to tweak it, then we can have that discussion. One change at a time. Maybe by then we'll have someway to do a trueskill kinda thing.
M57 wrote:M57 wrote:I'd hate to see the 1 point per net game paradigm get mauled by arbitrary add-ons.. So, 10 points for #1, huh? Regardless of whether you have 1100 or 3000 points to get there? I assume you also want to have minimums. Effectively, what you're talking about is adding Option I to the existing system.
FWIW, I'd rather add I-points to the existing CP system (or some version of the existing alla korrun's suggestion) rather than leave it alone. Even though this would make things even more convoluted, at least it adds an egalitarian element. In the sense that the 1 point per game ethos would become muddled, it doesn't matter as much what the multiplier/divisor becomes. A more important consideration would be balancing the two components.
Can folks perhaps weigh in on this thread?
Firstly, would you endorse ICPs at all? ..and if so, would you prefer them straight up alone (1 point per 20 GR), or would you like them 'enhanced' with something along the lines of korrun's suggestions?
I'd be fine with adding Option I to existing CP system, but would prefer to see something simpler to add to the top spots. I would prefer a flat number rather than a percentage, so as to balance out the number of points given to less popular boards (currently the ICP system would give 100 ICP to #1 on WGWF and 15 ICP to #1 on Battle of Waterloo).
Altogether though, that seems to be less popular.
To take the compliment of M57's stance, I would rather see a straight 1 ICP per 20 GR system than leave it how it is now.
I like the ICP system as proposed. IMO, it makes sense to give some kind of a bonus to the top spots, because I think it will increase competition for those levels. It does complicate things a bit, but overall I don't think it will be any more complicated than CP currently is.
I think the % option by AfroDaby makes sense. Something like
#1 = +10%
#2 = +5%
#3 = + 2.5%
Ozyman wrote:I like the ICP system as proposed. IMO, it makes sense to give some kind of a bonus to the top spots, because I think it will increase competition for those levels. It does complicate things a bit, but overall I don't think it will be any more complicated than CP currently is.
I think the % option by AfroDaby makes sense. Something like
#1 = +10%
#2 = +5%
#3 = + 2.5%
Sounds like consensus is moving toward ICP+. My vote is still in the ICP only column, but if you're going to add bonuses for the top spots..
I prefer % because it effectively limits or even prevents bonuses for low scoring board leaders.
Assuming the formula ROUNDDOWN(GR-1000)/20, a top score of 1200 would be necessary to get 1 extra CP at 10% (200/10=20 = 1 point), and for the #2 spot to get a point they'd have to have 1400. Seems a bit stingy, don'tyathink? Even though I'm not a fan, if you're gonna bother messing with the system, have it make a difference..
Here's how 40 20 10% breaks down by place.
GR | 1st | 2nd | 3rd |
3000 | 40 | 20 | 10 |
2500 | 30 | 15 | 7 |
2000 | 20 | 10 | 5 |
1500 | 10 | 5 | 2 |
1250 | 5 | 2 | 1 |
Sample board..
#1 player with 1750 GR gets 37 + 15 (bonus) = 52 ICPs
#2 player with 1500 GR gets 25 + 5 (bonus) = 30 ICPs
#3 player with 1450 GR gets 22 + 1 (bonus) = 23
#4 player with 1400 BG gets 20 + 0 (bonus) = 20
It takes a score of 1050 for the #1 player to get a single bonus point, and every 50 GRs accumulated with net an additional point.
Come to think of it, you could add 4th and 5th places at 5 and 2.5% respectively.. It's not much but it's worth a point or two once you're in the 2000 range. 1,2,3 seems about right.
Those #s seem a bit high to me for the bonus, but I guess it's just a matter of opinion. I think 20, 10, 5 is enough.
I did invention & WGWF with 40/20/10 and with 20/10/5:
https://docs.google.com/spreadsheets/d/1ScNMYxTV2rvoLFsOuN-_V1bF4wiZgBMFSxrAgPAJ8RU/edit?usp=sharing
It's globally editable, so feel free to play around, but leave it in good shape for the next person.
EDIT : On second thought after seeing the numbers, I agree 40/20/10/5/2.5 is probably a good way to distribute. On Invention for example, 1st & 2nd are pretty close, and the 40/20/... split means 1st gets about 11 extra CPI for being in 1st. Seems about right.
Like WT and M57, I think I prefer the non-additive CPI as well. It has a nice simple 'what you see is what you get' appeal to it.