Guild Wars 2

So as I understand it a couple of things will occur:

1. Scores will reset at end of this week
2. Matchup next week will be decided by the rankings this week, before they are reset

So essentially the three tier 1 servers by rankings at the end of this week will still fight each other next week as if there was no reset. But FOLLOWING next week, ranks will start anew.

How will tier 1 be defined after next week? Will it be the three servers who win by the widest margin? That wouldn’t really be the three top servers, just the three servers with the widest margin between the, and their competition.

However, if the new rankings take into account previous tiers to decide score, then the rankings aren’t really going to be reset completely.

So are we to expect several weeks of the tiers being all weird? Tier 1 being made up of the three servers who were furthest from their competitors (but not necessarily the top three servers). How long do people think it will take for things to settle?

Why are they using this frustrating method to balance the rankings when things could be accelerated from weeks to days by adjusting matches to much short length for a period.

It is likely some matches will be lop sided week after next meaning unfun, boring gameplay as two weak servers get smashed by one of the bandwagons.

If it is going to work the way I expect, they will be lopsided for much longer than a week. With week long matches it will likely take at least a month or two for things to settle.

drwookie.6391:

So as I understand it a couple of things will occur:

1. Scores will reset at end of this week
2. Matchup next week will be decided by the rankings this week, before they are reset

So essentially the three tier 1 servers by rankings at the end of this week will still fight each other next week as if there was no reset. But FOLLOWING next week, ranks will start anew.

How will tier 1 be defined after next week? Will it be the three servers who win by the widest margin? That wouldn’t really be the three top servers, just the three servers with the widest margin between the, and their competition.

However, if the new rankings take into account previous tiers to decide score, then the rankings aren’t really going to be reset completely.

So are we to expect several weeks of the tiers being all weird? Tier 1 being made up of the three servers who were furthest from their competitors (but not necessarily the top three servers). How long do people think it will take for things to settle?

This is exactly my reading of it.

Top three margin of victory servers go directly to T1. So, the top server from T8 in theory could jump all the way to T1, 1st place.

If this is the case, it will take 3 weeks for things to resemble any reasonable estimation of which servers have the highest population and therefore should be fighting each other.

Week 1, everyone is equal.

Week 2, weird matchups. High pop servers destroy competition.

Week 3, it’s close to normal, but with a few ‘odd’ placements, but they show their stripes.

So, by week 4 we have mostly everything settled except some servers on the cusp of where they should be.

That’s how I read it, anyways, but would love clarification from an official source.

Itll be like the first few weeks after launch. Servers will eventually stabilize again and since there are no free transfers, servers will not suddenly grow super strong again and mess up the tiers.

Don’t forget they’re also changing the math for how ratings are calculated. So take Kaineng for example, a T8 server with a sudden influx of transfers dominated T8 for a few weeks before moving to T7, and then dominated each tier only moving up 1 per week.

The goal of the new math is to make a server like that move through the tiers much faster. I’ll keep my fingers crossed and hope that it works and things don’t take a whole month to stabilize.

Vilkata.4725:

Don’t forget they’re also changing the math for how ratings are calculated. So take Kaineng for example, a T8 server with a sudden influx of transfers dominated T8 for a few weeks before moving to T7, and then dominated each tier only moving up 1 per week.

The goal of the new math is to make a server like that move through the tiers much faster. I’ll keep my fingers crossed and hope that it works and things don’t take a whole month to stabilize.

ok, sure that might make sense for Kaineng, but what about winning servers in top tiers that have sustained victory records, those dont count for anything? It’s just not fair.

So, even MORE emphasis on margin of victory?

rchu.8945:

Vilkata.4725:

Don’t forget they’re also changing the math for how ratings are calculated. So take Kaineng for example, a T8 server with a sudden influx of transfers dominated T8 for a few weeks before moving to T7, and then dominated each tier only moving up 1 per week.

The goal of the new math is to make a server like that move through the tiers much faster. I’ll keep my fingers crossed and hope that it works and things don’t take a whole month to stabilize.

ok, sure that might make sense for Kaineng, but what about winning servers in top tiers that have sustained victory records, those dont count for anything? It’s just not fair.

Considering the top tiers will start at the top, it’s entirely possible they may stay there, you will likely see the most volatility in T5-T2

tkalamba.2541:

rchu.8945:

Vilkata.4725:

Don’t forget they’re also changing the math for how ratings are calculated. So take Kaineng for example, a T8 server with a sudden influx of transfers dominated T8 for a few weeks before moving to T7, and then dominated each tier only moving up 1 per week.

The goal of the new math is to make a server like that move through the tiers much faster. I’ll keep my fingers crossed and hope that it works and things don’t take a whole month to stabilize.

ok, sure that might make sense for Kaineng, but what about winning servers in top tiers that have sustained victory records, those dont count for anything? It’s just not fair.

Considering the top tiers will start at the top, it’s entirely possible they may stay there, you will likely see the most volatility in T5-T2

Nope. It’s a complete reset. Those top 3 servers next week will be rated based on how they do against each other, creating a new and original Glicko rating. If they keep the fight close, ALL of their ratings will end up on the mid to low end, putting them in the lower half of the ranks, and into T4 or 5 or even LOWER.

T1 wins do not get more points than T8 wins that first week.

I see it as you guys, is like having your house in flames and trying to fixing it throwing some fuel, i can see some qq on the forums in the next weeks.

The only solution i see rigth now is to make the first five or six encounters to last 1 or 2 days instead a week.

The way I see it, explained by an example…

Next week’s tier, is fully based on the score of what the servers will have friday at reset, with the old system. I take the numbers from http://mos.millenium.org/matchups#NA as example)

T1 = JQ, SoR & SoS
T2 = BG, TC & FA
T3 = Kai, Mag, DB
T4 = SI, YB, CD
T5 = EB, IoJ, DR
…

All servers will have ‘Old ranking = 0’ in the first week.

Assume the following scores at the end of week 1.
T1 = JQ = 90k, SoR = 80k & SoS = 70k
T2 = BG = 51k, TC = 121k & FA = 71k
T3 = Kai = 92k, Mag = 82k & DB = 72k
T4 = SI = 103k, YB = 83k, CD = 63k
T5 = EB = 84k, IoJ = 94k, DR = 74k

Then week 2 will look like this
T1 = TC, SI & IoJ
T2 = EB, YB & Kai
T3 = JQ, Mag, DB
T4 = SoR, FA, CD
T5 = SoS, DR & BG.

So, the tiers in week 2 will be 100% based on the margin of victory in the first week. Because everyone’s ranking in the fiorst week is the same, there are no higher or lower ranked servers in your tier, so this factor is not taken into account.

Week 3 ups and downs will depend on the same as now, the margin of victory and the ranking of the serves in your tier compared to your own ranking.

This is the week were the old top servers are going to have to proof themselves.

And it will be pure, blind luck for the other servers if they will be pitted against a top-3 server or not in this week.

Result?
Lower servers will be tempted to do what SoR did last week and manipulate the rankings at the end of week 2. Just a few points up or down can make a huge difference: it will decide if you are in a tier with a top-3 server in the 3rd week or not. And the result in week 3 is very important, they will determine where your server will be in week 4 much, much more than the results determine rankking snow, due to the reset.

Tallis.5607:

The way I see it, explained by an example…

Next week’s tier, is fully based on the score of what the servers will have friday at reset, with the old system. I take the numbers from http://mos.millenium.org/matchups#NA as example)

T1 = JQ, SoR & SoS
T2 = BG, TC & FA
T3 = Kai, Mag, DB
T4 = SI, YB, CD
T5 = EB, IoJ, DR
…

All servers will have ‘Old ranking = 0’ in the first week.

Assume the following scores at the end of week 1.
T1 = JQ = 90k, SoR = 80k & SoS = 70k
T2 = BG = 51k, TC = 121k & FA = 71k
T3 = Kai = 92k, Mag = 82k & DB = 72k
T4 = SI = 103k, YB = 83k, CD = 63k
T5 = EB = 84k, IoJ = 94k, DR = 74k

Then week 2 will look like this
T1 = TC, SI & IoJ
T2 = EB, YB & Kai
T3 = JQ, Mag, DB
T4 = SoR, FA, CD
T5 = SoS, DR & BG.

So, the tiers in week 2 will be 100% based on the margin of victory in the first week. Because everyone’s ranking in the fiorst week is the same, there are no higher or lower ranked servers in your tier, so this factor is not taken into account.

Week 3 ups and downs will depend on the same as now, the margin of victory and the ranking of the serves in your tier compared to your own ranking.

This is the week were the old top servers are going to have to proof themselves.

And it will be pure, blind luck for the other servers if they will be pitted against a top-3 server or not in this week.

Result?
Lower servers will be tempted to do what SoR did last week and manipulate the rankings at the end of week 2. Just a few points up or down can make a huge difference: it will decide if you are in a tier with a top-3 server in the 3rd week or not. And the result in week 3 is very important, they will determine where your server will be in week 4 much, much more than the results determine rankking snow, due to the reset.

What this guy said. His rankings are just examples, but his maths and theory is spot une.

Jude.5408:

The only solution i see rigth now is to make the first five or six encounters to last 1 or 2 days instead a week.

This would not be a solution as diff servers have diff numbers on diff days of week, some have more ppl fri-sun and less ppl mon-fri others have roughly the same everyday

kingkillar.6714:

Jude.5408:

The only solution i see rigth now is to make the first five or six encounters to last 1 or 2 days instead a week.

This would not be a solution as diff servers have diff numbers on diff days of week, some have more ppl fri-sun and less ppl mon-fri others have roughly the same everyday

Its a good deal better than spending 3/4 weeks getting smashed or being so dominate there is nothing to do

What might make it even more interesting would be to reset the brackets to 4 servers.

3 Server Borderlands….1 roving server in search of a borderland to conquer and call their own.

Tallis.5607:

The way I see it, explained by an example…

Next week’s tier, is fully based on the score of what the servers will have friday at reset, with the old system. I take the numbers from http://mos.millenium.org/matchups#NA as example)

T1 = JQ, SoR & SoS
T2 = BG, TC & FA
T3 = Kai, Mag, DB
T4 = SI, YB, CD
T5 = EB, IoJ, DR
…

All servers will have ‘Old ranking = 0’ in the first week.

Assume the following scores at the end of week 1.
T1 = JQ = 90k, SoR = 80k & SoS = 70k
T2 = BG = 51k, TC = 121k & FA = 71k
T3 = Kai = 92k, Mag = 82k & DB = 72k
T4 = SI = 103k, YB = 83k, CD = 63k
T5 = EB = 84k, IoJ = 94k, DR = 74k

Then week 2 will look like this
T1 = TC, SI & IoJ
T2 = EB, YB & Kai
T3 = JQ, Mag, DB
T4 = SoR, FA, CD
T5 = SoS, DR & BG.

So, the tiers in week 2 will be 100% based on the margin of victory in the first week. Because everyone’s ranking in the fiorst week is the same, there are no higher or lower ranked servers in your tier, so this factor is not taken into account.

Week 3 ups and downs will depend on the same as now, the margin of victory and the ranking of the serves in your tier compared to your own ranking.

This is the week were the old top servers are going to have to proof themselves.

And it will be pure, blind luck for the other servers if they will be pitted against a top-3 server or not in this week.

Result?
Lower servers will be tempted to do what SoR did last week and manipulate the rankings at the end of week 2. Just a few points up or down can make a huge difference: it will decide if you are in a tier with a top-3 server in the 3rd week or not. And the result in week 3 is very important, they will determine where your server will be in week 4 much, much more than the results determine rankking snow, due to the reset.

Your math is flawed, TC will never beat BG =P

If you end up in a match with a giant server and you are a tiny server, contact the leading guilds on that server and organize some GvG stuff and activities to keep the smaller server interested. We’re not monsters!

Oh wait… fangs. kitten

tkalamba.2541:

rchu.8945:

Vilkata.4725:

Don’t forget they’re also changing the math for how ratings are calculated. So take Kaineng for example, a T8 server with a sudden influx of transfers dominated T8 for a few weeks before moving to T7, and then dominated each tier only moving up 1 per week.

The goal of the new math is to make a server like that move through the tiers much faster. I’ll keep my fingers crossed and hope that it works and things don’t take a whole month to stabilize.

ok, sure that might make sense for Kaineng, but what about winning servers in top tiers that have sustained victory records, those dont count for anything? It’s just not fair.

Considering the top tiers will start at the top, it’s entirely possible they may stay there, you will likely see the most volatility in T5-T2

This is incorrect. Someone has to lose the T1 match. That losing server will demolish the servers they get paired up with for the next month. This only complicates a ranking system that has almost worked itself back to where it should be.

I feel sorry for whoever has to play JQ/SoS/SoR in 2 weeks.