So about 1-2 years ago there was a huge debate about how to change CP calculations to be more equitable and recognize high achievers on popular boards who often are out of the points even being well over 1500 or even 2200 (WGWF).
A person who racks up 2200 points on a board can have 0 CP where someone with 3-4 wins on an unpopular board could have 10 or more.
Option K aka trueskill like system was the top vote. And, Tom offered to run it if someone else could figure out how to make it work. Over 1 year later? no one. nothing.
So why not re-open the conversation/voting/debate but no talking about Trueskill unless you're offering to program it yourself.
Below are the general options previously suggested, which would you prefer and why?
Option A:
+1 CP to every player over 1500 for every player over 1500 (current threshold).
It would make everyone over the mark "in the points" and would make 1st place on wgwf/Civil/etc. worth 120+ CP's for first place.
Option B
+ X% to each player over 1500 for each player over 1500.
Option C
Basic scale expansion extending the current threshold to 2500 or 3000 and pay out the top 20-30. (3000 would put the points above ALL current scores)
Option D
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,..
Option E
Logarithmic scale (example):
# of plays |
Top Rank CP |
Min GR for top rank |
# of ranks that earn CP |
0-10 |
10 |
1250 |
10 |
10-100 |
20 |
1500 |
20 |
100-1000 |
30 |
1750 |
30 |
1000-10000 |
40 |
2000 |
40 |
10000-100000 |
50 |
2250 |
50 |
100000-1000000 |
60 |
2500 |
60 |
Option F
A system for factoring in difficulty, such as:
CP x Difficulty factor x Popularity factor
Option G
The current CP system with a log modifier to weight it. A compromise between exponential growth & linear growth. So once x games are played or players played, it'd be x2 or x3, or x4, etc. Most boards keep the 20/1500/10 that we currently have. Popular boards get 40 CP. Very popular boards get 60 CP, ultra popular (Colossal, wgwf, etc.) gets 80CP,
Option H
Same as Option G but with a LOW log modifier to weight it. 1.5x, 2x, 2.5x
Option I
Simply calculate CPs by taking a player’s board rating and subtracting 1000.
If you have a 1750 rating on a board - that’s 750 CPs
negative ratings are thrown out - So no score is generated from a board ratings below 1000.
Option J
Expansion of Option I with Bonus CPs awarded to the top X players where each position gets an additional straight percentage of their current Rating for that board. For instance,
#1 = 100%, #2 = 95%, #3 = 90%, #4 = 85%, etc..
Search "Option" or "CP" if you want to see the hundreds of posts debating the topic.
Thanks INAT for posting this..
Just thought I'd organize my thoughts on these ideas in rough order of preference.
I. Because I'm biased when it comes to my ideas. That and because it's the most inclusive and egalitarian of the systems, awarding points for any level of achievement on any given board.
E. Same as above, but in a much more limited fashion
J. Why convolute a pure idea? Someone came up with the idea of making the point accrual threshold for an Option I system start at 1100. That's much more acceptable to me.
A. Because it at least increases the potential # of members who make it into "the club."
I'm opposed to all of the others equally because they do nothing in the way of making points more accessible to journeyman players.
Well, this is why I changed my mind about being a math major in college.
I think it would be easier for more people to weigh in if every option had an standard example, showing how it would affect the score of a make-believe player.
For example, Option A seems interesting, but it doesn't make sense to me. Do you get points only for people below your score who are also above 1500? Putting an example would help understand what that proposal is about, and I could say the same for some of the other options.
I think more clarity would be helpful, though I know it involves more work
Chele Nica wrote:For example, Option A seems interesting, but it doesn't make sense to me. Do you get points only for people below your score who are also above 1500? Putting an example would help understand what that proposal is about, and I could say the same for some of the other options.
Option A example.. Current standings on Invention. One of the most popular non-Risk boards.
CP's | ||
YeahBuddy | 15 | 2124 |
Andernut | 12 | 2027 |
YuriZ | 10 | 1823 |
Mongo | 8 | 1765 |
Cona Chris | 6 | 1747 |
bmasera | 4 | 1736 |
Insubordination | 3 | 1727 |
ivanfly | 2 | 1705 |
Avon | 1 | 1705 |
KillDawg | 0 | 1692 |
berickf | 0 | 1677 |
IJustWantURHalf | 0 | 1669 |
Willy Wonka | 0 | 1631 |
Mr Pink | 0 | 1571 |
timmy888 | 0 | 1553 |
beserker42 | 0 | 1511 |
My understanding is that Option A would give a single CP to the 7 players in the above list who currently have none. All players under 1500 would continue to receive 0 points.. With Option A, it's pretty clear that very few boards would qualify for additional CPs. No doubt it would have negligible if any impact on rankings.
But it also says that 1st place on some boards would get +120 points, so there must be something else to that proposal. It seems to imply everyone over 1500 gets points determined by how many players have passed the 1500 mark
Chele Nica wrote:But it also says that 1st place on some boards would get +120 points, so there must be something else to that proposal. It seems to imply everyone over 1500 gets points determined by how many players have passed the 1500 mark
Yeah - I'm not following where the +120 thing comes from given the description. Somehow 100 gets tacked on to the 20 for the player in first place.
M57 wrote:Chele Nica wrote:But it also says that 1st place on some boards would get +120 points, so there must be something else to that proposal. It seems to imply everyone over 1500 gets points determined by how many players have passed the 1500 mark
Yeah - I'm not following where the +120 thing comes from given the description. Somehow 100 gets tacked on to the 20 for the player in first place.
In wargear warfare there are currently 129 people above 1500. So the top player would get 129 CPs.
Korrun wrote:M57 wrote:Chele Nica wrote:But it also says that 1st place on some boards would get +120 points, so there must be something else to that proposal. It seems to imply everyone over 1500 gets points determined by how many players have passed the 1500 mark
Yeah - I'm not following where the +120 thing comes from given the description. Somehow 100 gets tacked on to the 20 for the player in first place.
In wargear warfare there are currently 129 people above 1500. So the top player would get 129 CPs.
..so the number #45 player on WGWF (1749) would get 84 CPs, and the #1 player of Spy vs Spy (1760) would get 4 points because there are only 4 people who have more than 1500 points on that board?
M57: One other question: With your I proposal, do you add all ratings to get the CP? IE, If someone has played three boards, and has 890, 1050, and 1500, then they would have 0 + 50 + 500 = 550 CP score?
Chele Nica wrote:M57: One other question: With your I proposal, do you add all ratings to get the CP? IE, If someone has played three boards, and has 890, 1050, and 1500, then they would have 0 + 50 + 500 = 550 CP score?
Yes. As I write this, you would have 1343 CPs
CPs | ||
Colossal Crusade | 1466 | 466 |
WarGear Warfare | 1162 | 162 |
Seven Redux | 1142 | 142 |
Infection | 1138 | 138 |
Dungeon Quest | 1096 | 96 |
King of the Mountains | 1089 | 89 |
Dungeon | 1056 | 56 |
Labyrinth: Cretan Minotaur | 1046 | 46 |
Antastic! | 1039 | 39 |
Random Mazes | 1034 | 34 |
Seven | 1026 | 26 |
Knight's Tour | 1022 | 22 |
Mobs of New York | 1018 | 18 |
Australian Risk | 1009 | 9 |
Total | 1343 |
The top advantage of Option I that comes to my mind is that newcomers are able to are able to collect CPs as soon as they begin to become reasonably successful on a board.
One on of those many threads the Option I CPs of some of the top players were calculated.. I'm not sure, but off hand the top players were in the 20 - 30+K range. Of course, if the numbers seem too high, they could be divided by 10 or 100 rounding appropriately, but I see no reason not to keep them as they are.
If dividing by 100 and rounding normally you would get no points for the bottom 7 boards listed above., and 1 point for Dungeon for 11 CPs total (or depending on the rounding rules no points on the bottom 10 boards and 1 point for Infection, Seven Redux and WGWF, for a total of 7 CPs). Any way you slice it, it's more CP production than the 3 points you currently now have under the current system.
Reasonable modifications suggested include raising the threshold to 1100, or 1250, etc. Personally, I think simple is better. The same effect can be achieved by dividing by 100 and always rounding down.
Thanks, this helps clarify. I still don't feel like I know enough to make a good vote on the options, and I wonder if actually doing the math for all these options in a google spreadsheet, for let's say 25 players who have played in 2015 (top 5 with current CP rating, top 5 with current tourney rating, top 5 score, top 5 H Rating an top 5 Team rating). I'm willing to put time into making calculations, if anyone else thinks it would be a good idea to try these out to see what they produce (and can lend me a hand), to help us figure out if they even make sense once we calculate the scores. I'm talking about 250 scores (25 players, 10 options each)
Most any of the proposed systems are bound to reorganize the top 25 players. Some will be happy, and some less so. Comparing systems at the top is bound to create some controversy, but speaking as someone in the lower echelons, I really don't have a strong opinion. However, I do about other considerations. What percentage of players are 'in the points' with these systems? - How do these systems treat popular boards vs. less played boards? How arbitrarily are they set up? Are the numbers pulled out of a hat (like the current system?) Are they scalable?
While I would agree that H-Ratings are a cool metric, because they are cumulative, I don't know how much value they have in this arena.
M57 wrote:Most any of the proposed systems are bound to reorganize the top 25 players. Some will be happy, and some less so. Comparing systems at the top is bound to create some controversy, but speaking as someone in the lower echelons, I really don't have a strong opinion. However, I do about other considerations. What percentage of players are 'in the points' with these systems? - How do these systems treat popular boards vs. less played boards? How arbitrarily are they set up? Are the numbers pulled out of a hat (like the current system?) Are they scalable?
While I would agree that H-Ratings are a cool metric, because they are cumulative, I don't know how much value they have in this arena.
I think you are selling yourself short with "lower echelon". You are currently in the top 3% of players with min 100 games based on CP.
While I was hunting for that data, I noticed that for players with >100 games played a GR of ~1000 puts somebody on page 29/42, so somewhere in the bottom 30%. The median (page 21/42) has players with scores of about 1200. For CP (>100 games) current median is 0 with those in the points starting on page 16 with person #480.
For >20 games played GR 1000 would put one on page 48/89, so about the middle if many less experienced players (but still stuck around to play for a bit) are tossed back in the mix. Page 22/89 puts people in the points here.
So it looks like a little more than 1/3rd of very experienced players are in the points in the current system, which drops to about 1/4 if we include are players I would consider to have a statistically relevant number of games (20). Just food for thought.
Gotta say, I was not a huge fan of option I before, but it has really grown on me.
Ozyman wrote:Gotta say, I was not a huge fan of option I before, but it has really grown on me.
Glad to hear that. I like it because it is simple. (Though if I could program Trueskill I would)
It still rewards play on multiple boards - and I think would encourage people to play on new or less popular boards as well as established boards. Right now a newbie who wanted to gain CP would be foolish to play Civil War where you need 2000 to get 1 CP - but that's a quite popular board so should/could hook newbies.
And since it doesn't reward anyone for being in the "top 10" it would open the door to creating a ranking board of current players only. (Which we can debate once we have it.)
When I look at my peers in GR versus CP, my CP peers do not strike much fear into me and I know I would dominate most of them in most games where the overall dice were anywhere on the spectrum from slightly against me to anywhere in my favour. I have not really looked at the different CP proposals, but, if players could be organized less arbitrarily and into closer skill 'sets' then I might think a bit more of CP rather then as it is now of just a treasure hunt for winning a few games on lesser played boards. I actually don't mind GR all that much as when I go down the list it does seem to group people together who are 'similar' in their skill level and most of the fluctuations are based on players changing their play style versus how disciplined they are to sticking to putting them self in as many optimum game situations as possible.
Do any of these CP models propose to actually organize players by their skill or would that be the trueskill model only, if at all? Again, I have not put much thought into the models, but, by my gut, they do seem to just emphasize redistributing a flawed CP system rather then to actually create anything that emphasizes and organizes players by skill... Or am I wrong to say that? At the end of the day, if I looked at a CP rank and was able to say 'these players around me can definitely stand their ground against me, and vice versa. Damn I get some epic battles against this group.' Then it would seem to be getting somewhere reasonable. As it is, the rank is completely arbitrary as to player skill and thus of little interest to me. I'd much rather look at GR and H-rank to determine a player's ability.
Interestingly, a quick perusal of the first page of GR and CP ranks shows that some low h-ranks make it into the CP ranks (16 of 30 ranks <71), but, their is a certain level of h-rank excellence to make it to the top of GR (all >/=71 with only one exception). You have to be good to have a high GR, whereas CP can be achieved by being good, sure, but also with great persistence... Until CP actually represents something great, and lists players who are actually great, then why should I care about ranking high on a list that can and does reward a persistent mediocrity over skill?
Amidon37 wrote:
(Though if I could program Trueskill I would)
Me too, but A TS system would replace the Global Ranking system. The current, and almost any of the proposed CP systems would sit on top of TS with some modifications, so changing the CP system doesn't preclude changing the GR system.
This may be a little OT, but it I think it bears revisiting for perspective's sake, especially if we're closing in on endorsing a replacement for the current GR system.
I think we should be asking ourselves what the point of having CPs is if the site already has a good GR (eventually TrueSkill-like?) system in place. I'll offer my short answer: to promote and recognize outstanding/exemplary play across a range of game-play types and boards. Right now, Ranks (General, Colonel, etc.) are based on CPs, but they could just as easily be based on GR. Ranks are relatively new to the site, and I think an argument can be made that there's some 'piling on' happening. I.e., now that there are Ranks, is there as much of a need for CP's? Why not just ties Ranks to GR?
Considering the above, and as I type this, I'm realizing that Option I makes the CP system somewhat transparent in that it doesn't fundamentally 'change' the nature of how GR impacts Rank, and I think I like this.
If this post made no sense I'd like to attach a disclaimer that I haven't finished my first cup of coffee this morning..