According to this, I should bandwagon onto CD because they have coverage at all times, thus people to play with on my wonky schedule. And are top of their tier.
Wow YB was the only server that never went into red at any time. I was hoping to see number of players though because that would very interesting. We just need Anet to hand over that info so we can all place blame on each other with an accurate chart.
Wow YB was the only server that never went into red at any time. I was hoping to see number of players though because that would very interesting. We just need Anet to hand over that info so we can all place blame on each other with an accurate chart.
?
The chart above shows Yak’s Bend going Red in 6 Timezones. Looks like Crystal Desert and Borlis Pass are the only servers with 24/7 coverage, in comparison to the rest of their current Tiers.
Wow YB was the only server that never went into red at any time. I was hoping to see number of players though because that would very interesting. We just need Anet to hand over that info so we can all place blame on each other with an accurate chart.
?
The chart above shows Yak’s Bend going Red in 6 Timezones. Looks like Crystal Desert and Borlis Pass are the only servers with 24/7 coverage, in comparison to the rest of their current Tiers.
Seems this is for current 24 hr period versus historical look over time? Those values may change drastically weekend versus weekday etc…..
Wow YB was the only server that never went into red at any time. I was hoping to see number of players though because that would very interesting. We just need Anet to hand over that info so we can all place blame on each other with an accurate chart.
?
The chart above shows Yak’s Bend going Red in 6 Timezones. Looks like Crystal Desert and Borlis Pass are the only servers with 24/7 coverage, in comparison to the rest of their current Tiers.
Seems this is for current 24 hr period versus historical look over time? Those values may change drastically weekend versus weekday etc…..
The ratings that you see are for the current week in the specific time periods.
For example, the ratings for 1 to 4 UTC are calculated by :
1. Getting the ratings for 1 to 4 UTC of the previous week
2. Then for every day in the week, combine the score that each server gain between 1 to 4 UTC
3. With the score calculated in #2, calculate the ratings for this week.
4. At reset (on friday), the rating for the week that just ended are permanently stored in the database
So, to answer your question, it is an historical look over time. It will take a few weeks for the ratings to stabilize and indicate the true strength of each server but once it does, it will indicate for each timezone the “true” strength of each server. It’s as reliable as the “true” strength that Anet gives us over there https://leaderboards.guildwars2.com/en/na/wvw
the numbers are quite misleading…
it is only true if the data consist of the same severs for several weeks
just like neural network which require lots of data to be accurate
i suggest putting “number of weeks” for that identical match up to give a better idea of the coverage between servers
the numbers are quite misleading…
it is only true if the data consist of the same severs for several weeks
just like neural network which require lots of data to be accurate
i suggest putting “number of weeks” for that identical match up to give a better idea of the coverage between servers
Nope they are not misleading.
Glicko2 is meant to give you the relative strength of every opponents. For example, if server A has a rating of 2000 and server B has a rating of 1500, you know that server A is stronger than server B even if they have never faced each other. It’s like that that it works in chess 2. You don’t need to face the world ranked player that has 2500 glicko rating to know that he will wipe your kitten in a competition.
As I said before, these numbers work as well as the official ratings and really represent the strength of each server in each timezone the way that Arena Net calculate them.
a lower tier server raise up to a higher tier, they happen to be capable of competing in that tier but not winning it or coming as 2nd. they will still gain glicko, no doubt.
however, looking at how your site present the data, it will show that server positive gain in all timezone, thus giving an impression that they have a full coverage comparing to other servers of the same tier even though it is just glicko gained compared to the server they replaced. do you get what i mean?
Ah ok, but don’t just look at the gain. You need to look at the number next to it too that represent the current rating.
By clicking on the rows, you can see the average PPT in each timezone. The average PPT is the best indication of who is winning each timezone in the current matchup. While the rating is the best indication we can get on your server compared to the 24 others.
Wow YB was the only server that never went into red at any time. I was hoping to see number of players though because that would very interesting. We just need Anet to hand over that info so we can all place blame on each other with an accurate chart.
?
The chart above shows Yak’s Bend going Red in 6 Timezones. Looks like Crystal Desert and Borlis Pass are the only servers with 24/7 coverage, in comparison to the rest of their current Tiers.
Seems this is for current 24 hr period versus historical look over time? Those values may change drastically weekend versus weekday etc…..
The ratings that you see are for the current week in the specific time periods.
For example, the ratings for 1 to 4 UTC are calculated by :
1. Getting the ratings for 1 to 4 UTC of the previous week
2. Then for every day in the week, combine the score that each server gain between 1 to 4 UTC
3. With the score calculated in #2, calculate the ratings for this week.
4. At reset (on friday), the rating for the week that just ended are permanently stored in the database
So, to answer your question, it is an historical look over time. It will take a few weeks for the ratings to stabilize and indicate the true strength of each server but once it does, it will indicate for each timezone the “true” strength of each server. It’s as reliable as the “true” strength that Anet gives us over there https://leaderboards.guildwars2.com/en/na/wvw
Thanks for the explanation. this makes sense on why someone saw all green and another not so much. They looked at different times and last week was screwy with holiday days thur/fri (and of course sat) for many here in the U.S.
