Discussion about the Coalition War system

Hello gentlemen (no ladies on the internet).

I have come to propose a change to the CW system, as current ranking system just isn't fair. Currently high ranked coalitions have a much harder time competing because of the following:

• CLNs can choose who and if they will fight;
  
• Having low competence makes your coalition more attractive in the competition, since you're naturally going for the best chances of winning;
  
• The system awards much less points to high ranked coalitions who win against low ranked ones;
  
• The density of high competence coalitions is much lower than the density of low competence coalitions, making high competence clns unable to compete fairly against other high competence clns.
  

That leads to a system where the winners are decided not by skill, but by availability of competence similar matches, which high ranked coalitions can't get due to low density of high competence clns. This can be seen in the current CLN standings, where low competence coalitions are in the top 5 not because they are surprisingly good, but because they have more options of play. I've been thinking about that and how to solve or how to get a system that would most closely represent and reward coalition skill, so I'd like to propose another, more simple ranking system - a ladder like system, removing CP and competence from the table altogether. Here's the rules I've come up with. Also, suggestions and criticism is highly welcome, as we should all work together towards a more just system.

• CLNs have to sign themselves into the ladder. Signed CLNs are placed on the last position. This way only the coalitions interested in participating in the competition will be in the ladder and inactive coalitions are cut.
  
  
• Duration of the ladder is 3 months, after that trophies are awarded (same way as they are right now) and the ladder is cleared. This is the way to preserve a coalition's legacy in the tournament. Perhaps trophies could be awarded to every coalition in the ladder but in a more generalized way: crimson medal for 1st place, gold medal for 2nd~4th places, silver medal for 5th~8th places, bronze medals for 9th~16th.
  
  
• If the lower placed coalition wins, they switch places. The basic mechanic of the ladder. Instead of using CP or competence to determine rank positions, the position itself is used.
  
  
• Coalitions can challenge another up to 5 positions above. This is to prevent drastic drops or drastic raises of coalitions.
  
  
• Coalitions can only decline 1 challenge per coalition per month. This is to prevent the situation where one coalition can hold the first position by only accepting challenges from the lowest ranked or from the one with the least chance of winning.
  
  
• Coalitions that don't play a CW for a week lose one position. Inactivity should not be awarded, and coalitions that don't challenge or aren't challenged shouldn't keep their positions.
  
  
• Challenges can only be sent to coalitions that have at least one officer or one leader online and two more other players. This is to prevent challenge spam and to encourage the challenges to be held in the same day.
  
  
• Leaders and Officers have 2 days to respond to challenges. To give the coalitions a little time to get their players. You shouldn't have to wait for a certain player, as inactivity should not be awarded.
  
  
• Accepted challenges should have 2 days to be played. This rule is to let coalitions adapt and schedule their games. Even if a challenge is accepted, if a game is not played and the coalition didn't play anyone in that week, they will lose one position.
  

This is my current idea. Let me hear what you guys think, let's get a solid, fair system going. I agree this will certainly be a lot of work, but would definitely make things more just for everyone involved, without the need of having CP, competence or any dark mathematics whatsoever.

The current system is perfect, you stated it yourself: "That leads to a system where the winners are decided not by skill, but by availability of fair matches"

Fair matches are the goal of every coalition, if you're clan can't find one, try using more low-skilled players instead of your best. In byzantia's case, you could do as I said above, as your cw with ULAP today seemed godly unfair considering ULAP is ULAP. But, that's just my two cents, they should have asked for a match to their skill. As I said, try using newer/lower ranked players, and you should be grand as gold.

كتب بواسطة Garde, 17.04.2012 at 20:44

The current system is perfect, you stated it yourself: "That leads to a system where the winners are decided not by skill, but by availability of fair matches"

Fair matches are the goal of every coalition, if you're clan can't find one, try using more low-skilled players instead of your best. In byzantia's case, you could do as I said above, as your cw with ULAP today seemed godly unfair considering ULAP is ULAP. But, that's just my two cents, they should have asked for a match to their skill. As I said, try using newer/lower ranked players, and you should be grand as gold.

I meant fair competence-wise. Even if we match ranks, we cannot get a good amount of CPs as the other coalition would if they fought a coalition with similar competence, thus even if the match is fair, the reward isn't and we have a much harder time staying competitive than low competence coalitions.

I love it!

The current system does need changing yes, just look at the current seasons table to know this!

Top 5 at this moment.

Hellenic Macedonians W4 L 6
Deutsche Koalition W6 L 1
Stalins Martians W6 - L 4
Byzantia W7 -L 0
Art of War W4 -L 6

I think from looking at the table the problem is clearly the competence factor. It seems harsh that 2/3/4 have better win loss ratio yet sit behind 1st on competence alone.

my Coalitions competence peaked at 1.24 for a while, a level that with respect to my clan overstated where we are really in our development and forced us to challenge very good coalitions with slim chance of victory for two reasons:

Firstly

At the time we were still in the midst of our 10 season games, it was worthless to us to challenge clans with lower competence (but that I consider our level) as the reward to risk factor in terms of competing for place's was not in our favour.
This is evident in our clan wars vs Hellenic Macedonians - we win + 37cp they win + 76 cp. The same can be said for our clan wars with Art of War - we win +33 they win + 78.(even though on both occasions they outranked us). This difference in reward is purely down to the competence factor and discourages us from competing against clans of our skill such as HM and AOW again untill thier competence level is similar to ours.

Secondly

Basically another reason to play clans that are more skilled was to lose competence. Knowing that the reward is based on competence encourages us to do this, but in the end makes the situation worse for higher clans.

What we are creating with the current system is a competition for the mid level clans where as Pulsereaction said 'high ranked coalitions have a much harder time competing'

Proposals =======

With regard to Pulse's suggestion here, with a few tweaks I think it is worth looking into. The idea of a tiers system/ ladder system will definitely improve the competitiveness factor. Not only will it match existing coalitions that are the same standard together creating 'fair' wars, the development of each and every clan will be dictated by skill and not with our situation by competence alone. Also having a ladder system will encourage more clans new and old to compete.

When starting the Martians it would have been nice to have been grouped with other untried clans, we were lucky having started with players that had been in clans before, if this had not been the case then I think motivation to play and compete would go after a few loses. The ladder system would for new clans be more appealing than the current system as they know that they will have fairer fight.

In short I agree with the proposal, but im not very good at working out the nitty gritty (or explaining myself)

What I would like to see if it is to be changed is a situation where a Mid-Level and active clan like ours can compete fairly against clans of our standard and not be forced to not compete due to there being no reward in doing so!!

I have to say, very well thought out criticism of the current CW system.

My only criticism is that your ladder system seems a little too aggressive/competitive in its current form. I understand that we want competition, but some of these rules make what could be enjoyable into something stressful. For example, "Leaders and Officers have 2 days to respond to challenges. To give the coalitions a little time to get their players. You shouldn't have to wait for a certain player, as inactivity should not be awarded." This seems a little too extreme for my tastes. I would call myself a pretty active player but even I can go a few days without signing on. Perhaps extend this time limit?

Besides that, I fall right into Nate's camp.

كتب بواسطة Acquiesce, 17.04.2012 at 21:40

I have to say, very well thought out criticism of the current CW system.

My only criticism is that your ladder system seems a little too aggressive/competitive in its current form. I understand that we want competition, but some of these rules make what could be enjoyable into something stressful. For example, "Leaders and Officers have 2 days to respond to challenges. To give the coalitions a little time to get their players. You shouldn't have to wait for a certain player, as inactivity should not be awarded." This seems a little too extreme for my tastes. I would call myself a pretty active player but even I can go a few days without signing on. Perhaps extend this time limit?

Besides that, I fall right into Nate's camp.

Sure, I think we should all give our suggestions to arrive at a common ground where everyone is satisfied. Perhaps challenges could only be made if your cln and theirs have the players for the CW online?

I already said few times how current CW system doesn't really work as planned, so I can only agree with this. But your proposal is a little bit too competitive.

"Coalitions that don't play a CW for a week lose one position."
"Leaders and Officers have 2 days to respond to challenges."
"Accepted challenges should have 2 days to be played"
This all should be a little more, maybe 2 weeks and 5 days or something like that.

And you should be able to decline few CW proposals, but not from the same coalition.

كتب بواسطة Caulerpa, 18.04.2012 at 01:37

I already said few times how current CW system doesn't really work as planned, so I can only agree with this. But your proposal is a little bit too competitive.

"Coalitions that don't play a CW for a week lose one position."
"Leaders and Officers have 2 days to respond to challenges."
"Accepted challenges should have 2 days to be played"
This all should be a little more, maybe 2 weeks and 5 days or something like that.

And you should be able to decline few CW proposals, but not from the same coalition.

Yeah, I'm pretty much online all the time so 1 week looks like an eternity to me, but these numbers are all arbitrary. I thought of an amount of time that would be significant in a 3 month period but would also be small enough to boost competitiveness. The best way to get the optimal numbers is by testing but we already have too many tournaments / competitions going on. I don't exactly know what is the best way to implement these in the game either.

Although i agree that the coalition system needs a change, i don't think this would be a good solution (no offence Pulse, i'm happy you come up with idea's).

Why should we invent the wheel again when there are so much other 'simulair' competitions on the internet? I've played in more leagues in the past and they all have their systems, of witch the most work very good. I will advice everyone to look at working systems with millions of players like: clanbase and ESL. Also, we don't need big changes, it can be way easyer so its easyer to implement.

Here is my proposal, take your time to read it (its simulair to existing league systems):
1. Remove competence and restore the old 'CP' system
2. Inactive coalitions needs to be removed from the coalitionlist and need to be placed in 'Hybernate'. When a coalition did not play a CW for 30days, it will automaticly be moved to the Hybernate mode and will not be shown in the coalition rankings.
3. Add a new feature "challanges" with the following options:
a. Coalition leaders and officers are allowed to arrange new challanges, or to accept/decline incomming challanges
b. When challanging a other coalition, all the 'gamerules' must be stated in the challange. Also there must be a date and time.
c. If someone is challanged, then there are 3 options: Accept, Change (date or rules) or Decline. Here is how it works:
- Accept: If you accept a challange, then everything is set. If one of the opponents doesn't show up on the scheduled date/time, the opponent wins by forfeit.
- Change: Challanges can be 'rechallanged' (changed):
- To propose a new date/time
- To propose a new rule/map
- Decline:
Rejecting a challange should only be possible in one of the following cases:
- The coalition war is scheduled within 5days from the challange date
- The opponent coalition has 100CP more or less then yours. You don't have to play against opponents that are far better or worse then your coalition
- If you already played against this opponent in the last 60days
- You are currently in Hybernate mode
- If you or your opponent proposed new rules or a date time for the 5th time and still doesn't have any agree with the opponent, then both side's can cancel the match. Canceling with this option should be limited though, so it can't be abused.
- If you decline a challange when you do not meet any of the requirements above, then you will be placed in Hybernate automaticly.
d. Once a match is accepted, its not possible to undo/reschedule this. Except when moving to hybernate mode.
4. Not responding to a challange within 7days, will automaticly result in a hybernate mode.
5. Going into Hybernate mode will cost the coalition -50CP. So activity is rewarded and you can't abuse the hybernate mode to decline incomming challanges.
6. Coalitions in hybernate mode can't respond on incomming challanges and can only challange themselves.
7. New coalitions will start in Hybernate mode without the CP loss, so they don't have to play directly when created.

Then also i have 2 points that would be nice to add, but not directly necessary for a good coalition league:
8. Show all the played matches on the league page, so everyone can see what coalitions played recently and what the result was. So i don't have to click on a coalition to see the last results. (I already proposed this before). Also (if my idea's are implemented), show 'upcomming matches' with all the agreed matches that are comming up. This way everyone can see what matches will be played and can maybe spectate them.
9. The CP you earn on winning or losing a CW, depens on the ranks of the players in the game (not on the CP of the coalition). This gives a opportunity for higher ranked coalitions to get more points.

Good points Hugo. I especially like idea about a window showing the results of recently played wars. However I think that your proposal is again over competitive, with just about every choice leading to the hibernation mode. And at least Pulse's proposal can be changed in the ways Caulerpa suggested to fix this problem.

Remember guys, the main problem with current system is not that there aren't enough clan wars or that there isn't enough competition. It's the fact that the top coalitions each season aren't really the top clans. When training clans are consistently outscoring their respective main clans, you know something is up.

My idea is not ment to solve any problem about inactivity, but to get a more fair competition. The days and/or ways when you go into hybernate mode can be changed, the one's i made where based on other leagues with a simulair system. However, not playing for more then 3months should get you out of the league for sure.

Actually, in Hugo's system, over-competition is not a problem since anyone who doesn't want to CW often and has a coalition purely for fun etc. can go to hybernate mode and nobody is forcing him anything in that case.

I like your idea, Hugo, much simpler than mine as it will not need to revamp the system so much. And would make lower ranked players participate more in the competition. I just don't understand this, Hugo:

اقتبس:

- If you already played against this opponent in the last 60days
- You are currently in Hybernate mode

If you play someone you've played before you go into hybernate?

BTW, Caul, my idea of hybernation is just not participating in the ladder, which is a bit extreme as it's "either you're in or you're not".

كتب بواسطة Caulerpa, 18.04.2012 at 06:44

Actually, in Hugo's system, over-competition is not a problem since anyone who doesn't want to CW often and has a coalition purely for fun etc. can go to hybernate mode and nobody is forcing him anything in that case.

Yes I read it again and I see your point. And in terms of realism I think Hugo's would probably be easier to implement as it isn't a total revamp of the current system. Just a remedy for the main problem.

كتب بواسطة notserral, 18.04.2012 at 10:52

I just don't understand this, Hugo:

اقتبس:

- If you already played against this opponent in the last 60days
- You are currently in Hybernate mode

If you play someone you've played before you go into hybernate?

No. These where 2 points when you are allowed to 'decline' a incomming challange. So, you can decline it when you recently played the opponent (so i will not spam invitations to the same coalitions). Or, if you are in hybernate mode, you can automaticly decline it without consequences (because you are already in hybernate mode).

I like your idea very much Hugo and would be much easier to implement. Can we use my dark mathematics formula for CP calculation?

اقتبس:

CP Gain = 50 * 2.71828^(-( RankDifference^2 / 50 )) + 10

The RankDifference variable is the difference between the sum of the ranks of players involved in the CW (= (rank player 1 + rank player 2 + rank player 3) - (rank player 4 + rank player 5 + rank player 6)). Here's a table exemplifying how rank difference would affect CP.

Rank Difference	CP Gain
0	60
1	57
2	50
3	40
4	31
5	22
6	17
7	13
8	11
9	11
10	10

The rating system is a other discussion i think. If we want something like this, it will take more time and it will be harder to implement. Thats also why i posted it as 'optional' feature. But if the rating system is going to change, then i prever to stick at the ELO system that has been discussed earlyer. This is how it works:

How many points you gain after a win or lose after a defeat depends on the rating difference between you and your opponent before the match. The ELO formula is very complicated (see last topic or WIKI), but the following easy examples will give you a good idea:
- Beating an opponent with the exact same rating gets you 60 points.
- The maximum amount of points you can win for a single victory is 120, but to get this many points you'll have to beat an opponent who is 700 or more points ahead of you in the ranking. You get 100 points for beating an opponent who is 200 points ahead.
- Beating lower ranked opponents gets you less than 60 points. If they were 200 points below you you only gain 20 points, if they were 300 points below you only get 10 points, and if they were 400 points below you get a mere 5 points.
- The number of points you can win and the number of points you can lose always add up to 120. So if you can only win 30 points with a victory, you can lose 90 points with a defeat.

However, you can also add players ranks with the total SP like this: players rank * 10.
For example:
3x rank 9 gives you a virtual +270. (3 * 9) * 10. (so the 270 will be counted + your total CP)
3x rank 10 gives you a virtual +300. (3 * 10) * 10 (so the 300 will be counted + your total CP)

I like the ELO idea, but how would it fit in your system? CP would be regulated by the ELO system?

If so then it would eventually fall on the same problems of the current system, where it doesn't take the ranks of the players into consideration for CP.

كتب بواسطة notserral, 18.04.2012 at 12:33

I like the ELO idea, but how would it fit in your system? CP would be regulated by the ELO system?

Exactly. See my first post

كتب بواسطة Hugosch, 18.04.2012 at 04:39

1. Remove competence and restore the old 'CP' system

However, coalition should start at a 1000CP rating, not 1500CP like it was.

This example will give you a good idea about how they work. I think its the best system there is for competitions like this.

Lets name the actual problem: Top CLNs get no CWs

But I dont see the problem so much in the CW system itself but in the CLN organisation.
A very small number of CLNs accumulates all the best and active players and their profound knowledge.
And they keep attracting and training the future top players to get even better.

Be honest if you were a random level <7 player in a non top CLN why would you want to fight them?
The match would most likely be over in 5 turns so you would hardly gain any experience but frustration.
In return the top players get frustrated aswell for a lack of competition and an overall small number
of CW games for them.

Other leagues might have nicer systems but they just work because their playerbase is very large
and there are more than a handful of teams that compete on the top level.

I dont think that it is wanted nor reasonable to split up all the players into equally skilled mini CLNs although
it would solve this particular problem. But the social aspect of the coalition system we have now would be changed
in a disastrous for a casual game like AW.

CLNs as we have them now are very important social unitys. This is a nice thing to have and to keep players motivated.
But just compare our situation to other games or sports even. It hardly makes any sense to have
a squad that is 4 times bigger than the usual number of players on one side of the team game (i.e. 3 in a usual 3on3 game).
And Im referring to Bite Me! here which is by far the smallest of the top coalitions.

Taking this into consideration I do not see the future of competitive AW play in the current CW system at all.
And the least I see it in changing cw systems back and forth.

-

So I could imagine a third but very different option maintaining the current system:
- gather top players
- play balanced 3v3 regardless of coalition
- endless fun

going further we could reconsider individual player ratings etc.
and an automatic match making system

You don't have to sign up for the ladder, he is just saying that if you do, got hard or go home bro