Guild Wars 2

Have you run a simulation where you include the bonus points but not the multiplier?

Very good and well thought-out scoring system. You have my +1

Exactly what we need to improve the fun in wvw!

Well done bro. The best plan that I have seen around to fix the imbalance in servers for WvW.

“Basically an interesting thing happened that I honestly didnt expect. Because the model was predicting that an overpopulated server would quickly take the map, the bonus had little effect in increasing the score of the other two servers. It did however lower the score of the oceanic server by almost 10k.”

“To the person saying that oceanics would be ‘worth less’. Thats hogwash. Currently they are worth much more than a primetime player, this brings their worth in line, makes it equal. And rewards effort against fighting advisories while still giving an advantage to a server with more people.”

Bullkitten, there I said it

You just went a made a system that “REDUCED” the Oceanic contribution, therefore making the current flawed system “Oceanic players are more important than North American players” and changed it to “North American Players are more important than Oceanic players” WITHOUT solving anything, Bravo

Your system flies against Anet’s wishes for a system that doesn’t penalize anyone regardless of Timezone.

“Server 1, is a popular US server with no significant help from outside its timezone, it gets really full during primetime
Server 2, is a less popular US server barely reaching queue at primetime, but has an oceanic presense.
Server 3, is an underpopulated server stuck in the matchup.”

Define your Oceanic Presence on Server 2

Your system takes a specific servers problem and then goes and dumps it on 2 other servers if it becomes a

Heavy NA, Moderate Oceanic
Moderate NA, Moderate Oceanic
Moderate NA, Heavy Oceanic

Pray Tell this doesn’t happen

Moderate NA, Average Oceanic
Heavy NA, Heavy Oceanic
Average NA, Moderate Oceanic

Server B gets stuffed right to hell and back with a Match up long debuff against it..

IF a group cant take a tower at any time and get the +15 Points for it, compared to the other servers, it isn’t going to be implemented, ANET DO NOT want to have people thinking they are Second Class Citizens

My Solution

-To come up with a system to balance and reinforce enjoyment of WvWvW
-Do it so that everyone’s efforts are Equal and Contributing at any time without any form of punishment/disadvantage

Analysis

-Current problem creating issue of night capping arises from the PPT system, or the potential for it to “Run Away” with itself if a server can get a +600 PPT for a certain particular amount of time if no resistance is met. If said situation occurs over a 8 hour period, 19200 points is awarded, If it repeats over consecutive days demoralization occurs as other servers in match up fall away in point score and unofficially declare the “Match Over”

Proposed solution

Combination of the old system with emphasis on 12 hour periods where the scoring system remains equal regardless of opposition met, population of other servers, Timezones.

Except you now Get Victory points

Victory Points are to be awarded after every 12 hours [(Set) Yes I’m stealing a idea from Tennis here, bear with me its 0040 in the morning for me and for lack of a better word] Who ever has the highest score with the current system wins the set with the points system being

Winner = 3 Point
2nd Place = 2 Point
Last Place = 1 Point

PPT accumulated scores reset back to 0 and the next set begins.

At the end of the 2 Weeks the Victory Points are added up with the winner being the one with the most points

Now

Server A has Good Daytime / Ok Night time
Server B has Good Daytime / Good Night time
Server C has Ok Daytime / Good Night time

So looking at this setup its fair to say

Servers A B will do well during the day, C might struggle
Server B and C will do Well during the Night, A Might Struggle

Match Plays out over 12 Sets (6 Days) with the following

Set 1 A 3 B 2 C 1
Set 2 A 1 B 3 C 2
Set 3 A 3 B 2 C 1
Set 4 A 1 B 2 C 3
Set 5 A 2 B 3 C 1
Set 6 A 1 B 3 C 2
Set 7 A 3 B 2 C 1
Set 8 A 1 B 2 C 3
Set 9 A 2 B 3 C 1
Set 10 A 1 B 2 C 3
Set 11 A 3 B 2 C 1
Set 12 A 1 B 2 C 3

Server A 22
Server B 28
Server C 22

So in this outcome as Team B has the Balanced Day/Night Ration compared to the other 2 servers after 12 Sets it would be declared the Winner right

But this is half the change

Look at it again

Set 1 A 3 B 2 C 1
Set 2 A 2 B 3 C 2
Set 3 A 3 B 2 C 1
Set 4 A 1 B 2 C 4
Set 5 A 2 B 3 C 1
Set 6 A 1 B 3 C 3
Set 7 A 4 B 2 C 2
Set 8 A 1 B 2 C 3
Set 9 A 3 B 3 C 1
Set 10 A 2 B 2 C 5
Set 11 A 3 B 2 C 1
Set 12 A 1 B 2 C 3

Server A 26
Server B 28
Server C 27

What the hell happened in this match up

Stone Mist Castle

Currently worth 35 + on its own

will now be worth 100 Points (Debatable)

Now your thinking Moose WTF Man??

here’s where it gets interesting

The emphasis is now on SM to attack it because its worth so many points, come on its a Awesome castle deserving to be God Like

At a Price

Notice how in some of the Match the loosing servers gained extra points? This is a result of a new Condition

“When 1st Ranked server holds SM and looses it to either 2nd or 3rd place server, these servers gain a bonus Victory Point as a “Humiliation” factor to the best server loosing to a lower ranked server"

“If first place re-captures Stone mist, off one of the other 2 servers, it is awarded no bonus, other than being able to have another chance at solidifying their lead”

“However a new Bonus Points is up for grabs it the other servers can take it”

“In case of two Drawing servers for first , the bonus point system is NOT awarded”

- In Summary

This scoring system in regards to the one week or 2 week matches that ANET will end up deciding on, emphasizes that each set is as important as the first ones, right down to the last one, with the Stone Mist Wild card, able to allow lower ranked servers the chance to claw back, should the leading server allow the to,

Ensuring

Everyone’s Efforts, regardless of timezone, location, “Primetime” matters to the final outcome

In Summary

Your System
-Actively DISADVANTAGES people due to prevent score blow out due to match up server population disadvantages
-Includes a system to assist with servers lacking in a timezone to claw additional points
-Snowball games can still happen, Games can still run away

My system
-Ensures everyone contribution is equal at any time of the day, at anytime of the match
-Includes a system to assist with servers lacking in a timezone to claw additional points
-Promotes that even the last 2 Sets can still be just as important then the current over by Monday system

@OP Really nice system. But I think people are not understanding it. Try to explain it better, maybe using smaller numbers (max population 10, number of cap points 6, or something) and then scale it up to show how that would affect real servers. Also try to remove referances from server names and timezones (Server A, B, C with 100%, 50%, 30% of maximum number of players is just fine), as ppl tend to get stupid :P when they hear about those (; Most of the complaints about your system I read here are about things that don’t affect your model (if I understood it correctly)…

+1 for this whole thread.

If ANET could just acknowledge this idea I would be so happy…

@Moose
The VP idea is good, but easy to abuse due to the SM extra VP.

Let’s say A and C ignored wvw, so they randomly got second or third place. Score at set 11 would be A 16/B 33/C 16.

Start set 12, a group of C infiltrates B. C then proceeds to go full power on SM and takes it, then proceeds to clean up of B invaders except for that infiltrated team. +1 VP for C.
C then lets the infiltrated Bs to capture SM. Since no one defends and they’re already inside, 10 minutes should be enough. No change in VP for B.
C now captures SM, also without much effort. +1 VP for C.
All C has to do is to put loads of siege in the throne room’s entrances to prevent B from cleaning SM. Almost no one would be able to take a corridor of death with 10 arrow carts, 5 ballistas and a bunch of extra people.

Repeat until the score is A16/B33/C33. In 6 hours or so, C got up to B’s VP, and could be very close in the set score from points gotten on the borderlands plus the points from SM. And this is in a worst case scenario. In an actual fight, C could get even is 2 or 3 hours of SM ping-ponging.

Sounds like a very well thought out system, I find that if the matches have 1 team so far ahead the other 2 stop playing which means less enjoyment for everyone involved

What I ment to address that you get more points for holding something than intentionally losing it, to recapture it later.

This certainly needs to be addressed. At the moment the zerg just runs around farming karma by capturing towers then leaving them to be lost before coming back and recapturing, and repeat ad nauseum. There’s little short-term incentive to actually stay and guard.

So what’s to stop the Oceanic server from gaming the system? If they take everything then most of them exit WvW and sit in Lion’s Arch, letting a small force defend, they would get a normal point score, wouldn’t they? But if they’re threatened, they just flood the map temporarily.

Secondly, if I’m understanding correctly, bonus points are scaled with the population modifier that is active when the structure is taken. This seems to discourage upgrading most structures and to let them fall back to equilibrium shortly before your server’s primetime in order to lower your opponent’s bonus points.

Also, without seeing all the details of the model, I’m not sure how close it matches reality. On Darkhaven, since we have a heavy US Pacific time presence as well as a moderate East Asian presence, we tend to start taking everything between 9PM and 11PM Pacific time. Our score goes from average (around +250) toward a blowout at almost +500 within the span of 2 hours.

What I would love to see is a way that games are about a balanced ranking system too.

Before IoJ reached Tier-1, we had 400,000 points in a week. The other two servers completely lost. It was extremely easy.

Right now we are fighting the fight of our lives in Tier-1 but I hate how the ranking is point oriented to the point that while we are fighting in the top 3 spots, each with around 60 – 70K a piece and the difference between 1st and 3rd being 1000 – 1500 points, where a 2000 point gap makes a REAL DIFFERENCE….

…I am in fear that a T-2 Server will simply wipe out 2 – 3 other servers and get 200 – 400K points and advance into a T-1 match, even though Tier 1 matches are exponentially tougher than T2 matches.

Draygo is credit to team. But you forgot one important factor.

Grizwhirl

But seriously you worked really hard on that and it looks like a great system. It was really disheartening to log off (see: power surge kick me out of) wuvwuv Friday night at a near even match only to be down an insurmountable amount 8 hours later. Because Australia.

Superb post.

They need to do something about night capping because people waking up and realizing no matter how well they do in their primetime all their work will be undone and the lead server establish an insurmountable point lead while they are asleep kills participation in Wuvwuv, leading to the series of blowouts we’ve been seeing in every tier since wuvwuv released.

I also like the bonus system: right now, the game downright encourages the two weaker servers to fight each other: those are the easiest points to obtain and their opponent won’t have the orbs and you make it more likely you’ll end up in second instead of third place. Points held by the ‘winning’ team absolutely should be worth more to capture to encourage them to work together.

Implement both of your suggestions and switch the orb and outmanned bonuses, and we may be able to salvage wuvwuv yet. After transfers are closed, of course.

Servers are getting more organized so I can seriously see a point where server A is up by an amount and organizes it such that they don’t login to wvw so the multiplier kicks in and the opposing teams can’t score enough to win.

Snido.4827:

Servers are getting more organized so I can seriously see a point where server A is up by an amount and organizes it such that they don’t login to wvw so the multiplier kicks in and the opposing teams can’t score enough to win.

I’d love to see modeling for the bonus points for capturing held objectives without modification of points gained for population disparities.

What if the modeling was set so the bonus for capturing a held objectives was 90% of the points rewarded for having held that objective?

Some great research being done. It would be nice if ArenaNet could provide the OP with some real numbers (and player expectations for Supply/Tower/Keep captures) to plug into his model for a more accurate picture.

Let’s face it. WvW imbalances are getting out of hand and making the “end game” feature of WvW not fun and useless majority of the week. Prime example would be the match ups Gate of Madness has faced the past two weeks (or more). Player morale should be diminished on the battlefield – not by scoring system numbers.

I like the idea of keeps giving more points the longer you hold them. On some maps you see a giant zerg of 70 or so just doing a circuit of the map, taking everything they see but making no attempt to defend it. You just take back everything they take after they leave, which is kind of boring since it’s more banging on doors.

And to Moose, his model used oceanic players on server 2 as an example. If the timezones are reversed it’s the same outcome. I don’t think you’re understanding his point.

Draygo can you run a scenario where a team takes most of the map then logs out ftw? Near end of end of match?
How about abandoning battlegrounds and stacking others? (Low pop server?)

Snido.4827:

Servers are getting more organized so I can seriously see a point where server A is up by an amount and organizes it such that they don’t login to wvw so the multiplier kicks in and the opposing teams can’t score enough to win.

An easy fix to this is to look at server population as opposed to WvW population.

Or, heck, don’t apply the penalty to servers that are currently a certain percentage behind.

Moose.6294:

You just went a made a system that “REDUCED” the Oceanic contribution, therefore making the current flawed system “Oceanic players are more important than North American players” and changed it to “North American Players are more important than Oceanic players” WITHOUT solving anything, Bravo

His system seems to be uncaring what time of day it is, its only based on duration objectives go without switching hands. Given you still get more points for ownership I think your objection is invalid.

His system doesn’t penalize oceanic players. It makes a players contribution (points earned) proportional to the difficulty of earning those points (enemy players contesting)

Currently a player who can play when no one else is around and can go easily cap stuff unopposed is vastly more rewarded for his time than someone who can only play when lots of enemies are around – and this goes against the stated goal of the GW2 dev team, of making sure that one player’s time is not worth more than another’s.

@grimpaw I can probably run that later and modify the last day for a complete logout, but if you go by an earlier graph i posted if someone was to log out most of the day it hurts that servers score. The multiplier is based on the ratio of your servers pop compaired to the other two servers combined, so one server being gone isnt going to kill the muliplier where server B or C couldnt catch you. Plus your leaving your longstanding holdings with probably large bonuses availible for the taking.

also abandoning one map and going to another is irrelevent as the multiplier is based on total wvw population, not total population in a single map.

The point about a weakness if a server mass jumps to LA is valid, i think i can mitigate that by taking the highest player total in the last hour as the population marker for the multiplier. So if you jump ship in entirety the other two servers can get some stuff back within that hour.

Couple of things i’m going to try, i’m going to apply mooses scoring system to the model and see the result as well as put mooses proposed population curves in and see how they matchup.

The other thing the model doesnt account for is what happens when servers intentionally target the winning server, one side effect the bonus system might have. The winning server is likely to have higher outstanding bonuses, encouraging attacks from the other two servers. Instead of where the middle server tends to get ganged up on now, by the third place server seeking to get into second and the first place server seeking to secure its lead.

This is fantastic. I’d love to see something like this happen to GW2’s WvW system.

The system as it is can be really, really fun (even if you aren’t winning), but only if:
1) You ignore the cumulative scoring
2) You’re not playing way outside your server’s strong timezone

Bumps, this is a fantastic idea!

It doesnt need to be rocket science.

https://forum-en.gw2archive.eu/forum/game/suggestions/Some-suggestions-to-fix-WvW/first#post503310

Needs some feedback. Simple and easy to implement and they could as easily do it, next down day.

None of those suggestions deal with out of control snowballing scores in any way.

While I agree with (some of) them, they are not something the topic of this thread is exploring.

I ran the numbers on a Score + Bonus system that someone here wanted to see. So removing the multiplier we see the same difference that we saw between the anet system and the multi only system. Basically the server with the best coverage increased its lead, while the second and third servers had hardly any gains in points.

I do think a multiplier is needed to tamp down on the score.

Here is a chart of all 4 types, only the adjusted + bonus model gives a catchable lead by thursday. note that these servers are considered ‘equal skill’ and in each one server 2 wins. Whats changing is the ‘how much’.

Draygo.9473:

None of those suggestions deal with out of control snowballing scores in any way.

While I agree with (some of) them, they are not something the topic of this thread is exploring.

Resetting the points tally per server and balancing server population via carefully and well designed thought out merges is not controlling snowballing scores? Ok…..so lets reinvent the wheel then and just confuse the hell out of everyone.

Why do I get the feeling here of the same thing that struck down Warhammer Onlines RvR is about to happen here? Lets rework the system 3 or 4 times…….and everyone gets sick of it and quits…..

Bumping for a great idea.

bump for anet to see. This is brilliant idea! population =/= skill. A half empty server with good loyal population and good strategy and discipline should always beat a full server with noob bandwagon jumpers and mindless zergs. That is what wvw should be about, not the other way around.

If this is implemented it should hopefully spread out the oceanic population as well.

Taemek.1602:

Draygo.9473:

None of those suggestions deal with out of control snowballing scores in any way.

While I agree with (some of) them, they are not something the topic of this thread is exploring.

Resetting the points tally per server and balancing server population via carefully and well designed thought out merges is not controlling snowballing scores? Ok…..so lets reinvent the wheel then and just confuse the hell out of everyone.

Why do I get the feeling here of the same thing that struck down Warhammer Onlines RvR is about to happen here? Lets rework the system 3 or 4 times…….and everyone gets sick of it and quits…..

No reason to take it personal, but none of your suggestions have any sort of anti-snowball mechanism in them. Your talking about prematch, not during match even in this post again.

Sure reversing the outmanned and orb buffs is a great idea that i fully support, it doesnt do anything to help prevent a score from snowballing. Even if anet follows your suggestions and just starts closing and merging wvw populations, you will still have unbalanced matchups. its plainly unavoidable unless you want to lump everyone together in 3 servers and give everyone really long queues.

Love the idea. Want to see implemented +1

@moose, I still plan on throwing your scoring system into the model just to see.

I’m also going to make a few tweaks to my scoring model and see if i can improve it.

JackJack.2478:

bump for anet to see. This is brilliant idea! population =/= skill. A half empty server with good loyal population and good strategy and discipline should always beat a full server with noob bandwagon jumpers and mindless zergs. That is what wvw should be about, not the other way around.

If this is implemented it should hopefully spread out the oceanic population as well.

posts like this always confuse me. the skill of individual players matters very little, considering how small the individual person commits to the war effort. therefore, it is the “skill of the server” that matters in who wins and who loses. the most important skill of any large body of fighters is organization. one of the most important parts of organization is getting people to show up when they are needed. therefore, a server with more people on at night is more organized, and hence, more “skilled” at a server level.

+10,000

This is simply a great idea. Addresses population AND value of objectives. ArenaNet needs to hire you.

Currently, I’m in a great match-up in Tier 1. But 1 of the servers has “mindless” zergs. While the other 2 have tactical zergs. It is a crappy way to play, but it works. You simply throw numbers at an object to suicide until you take it. Once the score drifts too far, people give up and it snowballs.

Then you get days of lameduck fighting with the mindless zergs being the dominate factor. So much better when it was even fighting with the tactical zergs. I hope I don’t need to spell it out about the difference.

This makes it so people are more likely to be excited about recovering the lead. They have a chance now.

As far as the earlier poster saying score is meaningless besides ranking us for the next match, I say bull. Because then we wouldn’t need to see the score. It is a psychological factor that affects play. It could have been hidden from view, and people could simply play. But it is there and glaring.

ArenaNet, Regina Bulbbureoorea (sorry for the butchering), Gail Gray, someone… notice this post and love the idea and put it in.

Draygo.

Put this post into the Suggestions Forum as well? Or link back to here?

It is relevant to both, and Anet needs to see it.

The Suggestion forum is where threads go to die.

You’ll have much more attention and discussion here.

Ok Moose I ran your system through my data and got the followin.

I couldnt estimate how often SM would flip in the matchup, so I have no idea how your comback mechanic would actually work, but here are the problems with the system you propose:

1) It emphisizes US and Oceanic players with a 12 hour split. European servers and players would be stuck playing in both sides of the split, not helping those servers at all.
If you split euro servers differently then US players playing on euro servers will be more valueable as they again, play both sides of the split.
2) It doesnt stop snowballing.
3) It has a comeback mechanic – however I find the following problems

By setting SM to +100 you make holding the entirety of EB far more important than any of the BL’s not only that the leading server only has to hold EB to maintain the lead both in points and preventing the other servers from utilizing the ‘comeback’ mechanic.
Each BL map has 145 points availible for the taking.
EB under your system would have 325 points.
425 points are availible on all 3 BL’s. So if the other two servers contest them they will split those points, not passing the 325 mark.

This makes EB far too important so the server thats able to take upgrade and camp EB is going to win, significantly hampering your comeback mechinism. It will also do horrible things to the queues for EB, as its so important everyone is going to queue for it. Anet reduced the amount of points SM was worth, probably for this reason. Later in the matchup, nothing else will matter other than SM, I dont think I will like that picture.

Anyway, here is the comparison on the base points, I cant quantify your comeback system because I really have no clue how much SM will flip when you make it as important as you made it in your post.

To make the comparison i compared % of the total score each server got.

Its been said before but is worth repeating – artificially modifying the scores goes directly against the server ranking system currently in place. The only reason the ranking system hasn’t been effective yet is because of all the server hopping going on. Take that away and I bet we’d begin to see more even matches.

Even with the server hopping we are slowly starting to see better matchups (at least on the US side). Tier 1 has been a ripper this week and tier 2 will be just as good once HoD drops out and SoR moves up.

Thats not correct Tzash, even if you took hopping away, things wont settle due to a lot of snowballing mechanics in the game. Servers will never be even with variant populations at variable times. Something anet cannot really control.

A scoring system that takes difficulty into account is nessisary for the long term health of the scoring system.

As demonstrated it wont adjust the outcomes, previously close matches stay close. Matches that get out of control get tamped a bit in order allow late comebacks. And if that server doesnt excercise it, at least they knew they had the chance.

The middle will never really find balance under the current scoring rules because its, win by a kitten ton, go up 5 spaces, get destroyed as badly as y ou won the previously go down 5ranks. it wont find balance till the snowballing score is takin care of.

Draygo.9473:

The middle will never really find balance under the current scoring rules because its, win by a kitten ton, go up 5 spaces, get destroyed as badly as y ou won the previously go down 5ranks. it wont find balance till the snowballing score is takin care of.

I’m sorry, but you are basing all your assumptions on a system where server transfers exist. There is no evidence to suggest that the server ranking system won’t do what its meant to do if server hopping was eliminated.

The proposed bonus point system is pretty much the same as lowering the points gained from any objective held for more than a certain amount of time.

Except that holding objectives at the end of the game is worth more and the bonus point system rewards the capturing faction, leaving the non-capturing faction further behind (thereby giving a huge advantage to the faction that comes online in the morning-early day and starts competing back against the night-capper faction).

And what’s your opinion of it.

The idea of the system is that it leaves the matchup undetermined for a longer period of time. So even if your server falls behind early, you can catch back up and steal the lead.

Right now under the current system, even in close matchups, you get at least several hours of lame-duck time. Where nothing you do matters to affect the outcome of the match. If someone is beating you by 2k, you might say wow, thats a close matchup, but there is nothing another server can do under the current system to take the lead in the last hour.

the problem gets much worse if some server gets a 40k-50k lead on the other two.

Does it advantage servers to spread their timezone coverage, sure. Does it break the game and let them win the matchup PvD, no.

bump
well thought out model, devs should take a close look.

In light of the new WvW Blog post I am bumping this thread for further discussion on alternate scoring systems.

I am oceanic and I see a lot of promise in the outlined system, I would like to pose a few minor questions.

Assuming the system was implemented.

How do you motivate players to participate in hours where they are outmanned, when the impression may become ,“stay out it lowers their possible score”.

Under your propsed systme what is the max 24hr score available if opponent/s chose to blackout 24 hrs prior to reset ?

I think in answered this somewhere, but sitting out is not an option if you want to get points or hold onto them.

Remember if you are ahead in the matchup and you have significant holdings not defending them is going to give huge bonus points to competing servers. And as long as the other TWO servers participate you sitting out intentionally only hurts your server.

The other side of the coin is the current system already guarantees winners, while the bonus points system alleviates that a little bit it isn’t going to allow a server to make up a 100k point deficit regardless. The system doesn’t try to make servers with larger populations disadvantaged but attempts to bring the total scoring closer together and make coverage gaps less important.