KSoD Squad Ranking Website - Discussion Topic

General topics, questions, suggestions, bugs, or anything Star Wars Commander related.
DarkRebel
Grand Admiral
Posts: 1000

Re: KSoD Squad Ranking Website - Discussion Topic

Post#211 » Mon Dec 03, 2018 8:25 am

Midge wrote:
SpaceDandy wrote:Given all the confidence the system has in knowing this squad, it ought to know they are, um, not good. No offense, but just objectivly not as good. Some data stabilizes quicker than other in data sets like this but should result in a stabilized understanding that this squad isn’t deserving of this rank - not conflating its confidence in the data with what that data is actually saying. Or at least better balancing between the number of data points and what those data points say. If the Baltimore Orioles played 1,062 games last season and the Red Sox played 162, at their respective winning percentages, one should be more confident how bad the O’s are and not overly diminish the Sox’s excellence based on fewer games played. Especially when the other squads/teams play closer to, per the analogy, 162 game schedule. This is how algorithms should get refined, seeing a result like this, understanding why and tweaking it.

Apologies to non baseball fans trying to follow that.


It is refined regularly as it gains confidence with other squads.

The problem here is people conflating TrueSkill with RANKING.

The system is not RANKING teams. It is giving them a True Skill score and a certainty based around that score. Then simply sorting squads based on the combined metric.

They are CERTAIN of this squads number with NO uncertainty ... and I would venture to say they are pretty correct in that assertion considering it is a terrible squad with a terrible True Skill score.

Everyone here is getting caught up in the ranking, when the ranking of a squad like AncientWarriors can fluctuate immensely because they allow their sigma to fluctuate quite often with a week off or more. That adds UNCERTAINTY ... and the system processes it as not being certain of where it should be, while it is still certain, given the amount of data it has, that the other squad sucks ass.

Also, the assumption that being ranked 435 is somehow taking someone's spot is kind of laughable. It is not difficult in the slightest to get ranked in the top 300 considering there are barely 300 legit squads playing at this point.

So a "legit" squad is a squad which wars twice a week. I get it.
To get to top 500, you don't have to win. Just get a bot to submit for war twice a week, don't need to do a single attack, and you'd get in top 500.
btw, there are 26K squads out there. If you have good attendance record (war twice a week), you can be in top 2%, even if you lose all your matches.


User avatar
Midge
Grand Admiral
Posts: 1070

Re: KSoD Squad Ranking Website - Discussion Topic

Post#212 » Mon Dec 03, 2018 8:51 am

DarkRebel wrote:
Midge wrote:
SpaceDandy wrote:Given all the confidence the system has in knowing this squad, it ought to know they are, um, not good. No offense, but just objectivly not as good. Some data stabilizes quicker than other in data sets like this but should result in a stabilized understanding that this squad isn’t deserving of this rank - not conflating its confidence in the data with what that data is actually saying. Or at least better balancing between the number of data points and what those data points say. If the Baltimore Orioles played 1,062 games last season and the Red Sox played 162, at their respective winning percentages, one should be more confident how bad the O’s are and not overly diminish the Sox’s excellence based on fewer games played. Especially when the other squads/teams play closer to, per the analogy, 162 game schedule. This is how algorithms should get refined, seeing a result like this, understanding why and tweaking it.

Apologies to non baseball fans trying to follow that.


It is refined regularly as it gains confidence with other squads.

The problem here is people conflating TrueSkill with RANKING.

The system is not RANKING teams. It is giving them a True Skill score and a certainty based around that score. Then simply sorting squads based on the combined metric.

They are CERTAIN of this squads number with NO uncertainty ... and I would venture to say they are pretty correct in that assertion considering it is a terrible squad with a terrible True Skill score.

Everyone here is getting caught up in the ranking, when the ranking of a squad like AncientWarriors can fluctuate immensely because they allow their sigma to fluctuate quite often with a week off or more. That adds UNCERTAINTY ... and the system processes it as not being certain of where it should be, while it is still certain, given the amount of data it has, that the other squad sucks ass.

Also, the assumption that being ranked 435 is somehow taking someone's spot is kind of laughable. It is not difficult in the slightest to get ranked in the top 300 considering there are barely 300 legit squads playing at this point.

So a "legit" squad is a squad which wars twice a week. I get it.
To get to top 500, you don't have to win. Just get a bot to submit for war twice a week, don't need to do a single attack, and you'd get in top 500.
btw, there are 26K squads out there. If you have good attendance record (war twice a week), you can be in top 2%, even if you lose all your matches.


Just because there are 26 thousand squads out there does not mean there are 500 good ones.

And no, you don't have to war twice a week. Just once every six days. If you think there are more than 300 legit squads out there, excellent. Maybe I was being unfair. There could be a good 350 or so.

But lets be honest, most squads ranked around 250 don't even know how to put an actual wave together.

There are not 26,000 active war squads out there, you would be naive to think that. But, you know, whatever works for making you throw a fit about not wanting to war a whole 60 times a year.
Leader of The Mob Syndicate

Home of TheMidnightMob, TheMorningMob, TheMiddayMob and TheMobWantsYou

Level X Rebel : Medal Maxed


DarkRebel
Grand Admiral
Posts: 1000

Re: KSoD Squad Ranking Website - Discussion Topic

Post#213 » Mon Dec 03, 2018 10:22 am

Midge wrote:Just because there are 26 thousand squads out there does not mean there are 500 good ones.

And no, you don't have to war twice a week. Just once every six days. If you think there are more than 300 legit squads out there, excellent. Maybe I was being unfair. There could be a good 350 or so.

But lets be honest, most squads ranked around 250 don't even know how to put an actual wave together.

There are not 26,000 active war squads out there, you would be naive to think that. But, you know, whatever works for making you throw a fit about not wanting to war a whole 60 times a year.

Putting in an actual wave, how to attack, base design... should already been factored into their win/lost record. No need to bring in irrelevant facts to cloud the issue. A good squad should win more than lose. There is a problem if a losing squad is ranked higher than a winning squad. That's the fundamental issue here. Squads that don't know how to win should be ranked at the bottom.

At the end of the day, any system that put attendance record above winning is simply flawed.

Just forget all the craps about mu, sigma, true skill points, ager ... for a moment, what is the single most important objective of a ranking system ? Isn't it supposed to rank squads on how good they are relative to others ?

(Now having said that, I am not actually against the concept of the ager ... but that needs to be adjusted such that the formula need to weigh winning much more than attendance record. Winning should be the number one criteria for any ranking system.)


User avatar
Midge
Grand Admiral
Posts: 1070

Re: KSoD Squad Ranking Website - Discussion Topic

Post#214 » Mon Dec 03, 2018 11:09 am

DarkRebel wrote:
Midge wrote:Just because there are 26 thousand squads out there does not mean there are 500 good ones.

And no, you don't have to war twice a week. Just once every six days. If you think there are more than 300 legit squads out there, excellent. Maybe I was being unfair. There could be a good 350 or so.

But lets be honest, most squads ranked around 250 don't even know how to put an actual wave together.

There are not 26,000 active war squads out there, you would be naive to think that. But, you know, whatever works for making you throw a fit about not wanting to war a whole 60 times a year.

Putting in an actual wave, how to attack, base design... should already been factored into their win/lost record. No need to bring in irrelevant facts to cloud the issue. A good squad should win more than lose. There is a problem if a losing squad is ranked higher than a winning squad. That's the fundamental issue here. Squads that don't know how to win should be ranked at the bottom.

At the end of the day, any system that put attendance record above winning is simply flawed.

Just forget all the craps about mu, sigma, true skill points, ager ... for a moment, what is the single most important objective of a ranking system ? Isn't it supposed to rank squads on how good they are relative to others ?

(Now having said that, I am not actually against the concept of the ager ... but that needs to be adjusted such that the formula need to weigh winning much more than attendance record. Winning should be the number one criteria for any ranking system.)


Except, it is EASILY possible for a squad to go 22-45 if they keep facing top 100 squads and for another to go 55-12 facing the next 100 squads ... and that certainly does not mean that the one that went 22-45 is worse than the one that went 55-12.

Again ... It actually is not a ranking system ... it is a system to figure out the correct rating for EACH squad, then it sorts them.

When you started talking about the adjustments to where winning matters more than attendance, etc., you're talking about fundamentally changing the algorithm, which more or less is a from scratch thing.

All that said ... anyone ... ever ... bitching over a squad being ranked 435 in a game that barely has maybe 300 real squads left ... is kind of silly.
Leader of The Mob Syndicate

Home of TheMidnightMob, TheMorningMob, TheMiddayMob and TheMobWantsYou

Level X Rebel : Medal Maxed


User avatar
Spock
Admiral
Posts: 794

Re: KSoD Squad Ranking Website - Discussion Topic

Post#215 » Thu Dec 06, 2018 9:01 pm

Just want to let everyone know the ager has been turned off again for this war outage.
Also note, a fix was put out today for squads stuck at 0 sigma. There should be no more squads trapped with 0 sigma now.
This will allow all squads that were stuck at 0 sigma the possibility of gaining true skill points again once wars are re-enabled. (Depending on their matchups of course)
Image

Leader of the imperial_enclave. Flagship of the enclave brotherhood of squads.
Always looking for active lvl10 commanders. Line app required.


User avatar
StrikerRebel
Commodore
Posts: 113

Re: KSoD Squad Ranking Website - Discussion Topic

Post#216 » Thu Dec 06, 2018 11:11 pm

Hi Spock, first thx for your awesome work, but one question, before last war, that we won, we had 29114 skill points, after that war we had 29094 skill points, how we lost it ?
Squad is ActiveDonatorYMW thx for response....
Leader of ActiveDonatorYMW (former TOP 50 Windows-Squad ActiveDonator)
We are on the way to get back in the Top50 and searching for fearless rebels fighting with us against the Empire.
Perks maxxed out at 50 and ranked in the Top200 now....


User avatar
Spock
Admiral
Posts: 794

Re: KSoD Squad Ranking Website - Discussion Topic

Post#217 » Fri Dec 07, 2018 3:05 am

StrikerRebel wrote:Hi Spock, first thx for your awesome work, but one question, before last war, that we won, we had 29114 skill points, after that war we had 29094 skill points, how we lost it ?
Squad is ActiveDonatorYMW thx for response....


The 0 sigma fix caused some adjustments in the rankings. It looks like your squads sigma was adjusted to 0.01 with the rest of the squads that were stuck with a locked sigma. In the short term you might have lost 20 skill points but you will now actually be able to gain more skill points (depending on your matches) instaed of being stuck and unable to gain skill points at all.
Image

Leader of the imperial_enclave. Flagship of the enclave brotherhood of squads.
Always looking for active lvl10 commanders. Line app required.


User avatar
ChiptheFish
Major General
Posts: 394

Re: KSoD Squad Ranking Website - Discussion Topic

Post#218 » Fri Dec 07, 2018 7:21 am

Spock wrote:
StrikerRebel wrote:Hi Spock, first thx for your awesome work, but one question, before last war, that we won, we had 29114 skill points, after that war we had 29094 skill points, how we lost it ?
Squad is ActiveDonatorYMW thx for response....


The 0 sigma fix caused some adjustments in the rankings. It looks like your squads sigma was adjusted to 0.01 with the rest of the squads that were stuck with a locked sigma. In the short term you might have lost 20 skill points but you will now actually be able to gain more skill points (depending on your matches) instaed of being stuck and unable to gain skill points at all.

Spock, first of all, huge thanks for the work you guys do. Big fan.
Is our squad an example of getting locked in a sigma rut? FriskyEwoks:
http://ksod.us/squad/074ce1e7-ad0c-11e5 ... df/history
Even with a string of wins, we seem to be pegged. Thanks for the comments!
Chip the Fish, Rebel: HQ Prestige, Medals: 28K, Battle Rank: 14(Android)
FriskyEwoks: Level 50 Squad - Join us!


DarkRebel
Grand Admiral
Posts: 1000

Re: KSoD Squad Ranking Website - Discussion Topic

Post#219 » Fri Dec 07, 2018 9:08 am

Midge wrote:
DarkRebel wrote:
Midge wrote:Just because there are 26 thousand squads out there does not mean there are 500 good ones.

And no, you don't have to war twice a week. Just once every six days. If you think there are more than 300 legit squads out there, excellent. Maybe I was being unfair. There could be a good 350 or so.

But lets be honest, most squads ranked around 250 don't even know how to put an actual wave together.

There are not 26,000 active war squads out there, you would be naive to think that. But, you know, whatever works for making you throw a fit about not wanting to war a whole 60 times a year.

Putting in an actual wave, how to attack, base design... should already been factored into their win/lost record. No need to bring in irrelevant facts to cloud the issue. A good squad should win more than lose. There is a problem if a losing squad is ranked higher than a winning squad. That's the fundamental issue here. Squads that don't know how to win should be ranked at the bottom.

At the end of the day, any system that put attendance record above winning is simply flawed.

Just forget all the craps about mu, sigma, true skill points, ager ... for a moment, what is the single most important objective of a ranking system ? Isn't it supposed to rank squads on how good they are relative to others ?

(Now having said that, I am not actually against the concept of the ager ... but that needs to be adjusted such that the formula need to weigh winning much more than attendance record. Winning should be the number one criteria for any ranking system.)


Except, it is EASILY possible for a squad to go 22-45 if they keep facing top 100 squads and for another to go 55-12 facing the next 100 squads ... and that certainly does not mean that the one that went 22-45 is worse than the one that went 55-12.

Again ... It actually is not a ranking system ... it is a system to figure out the correct rating for EACH squad, then it sorts them.

When you started talking about the adjustments to where winning matters more than attendance, etc., you're talking about fundamentally changing the algorithm, which more or less is a from scratch thing.

All that said ... anyone ... ever ... bitching over a squad being ranked 435 in a game that barely has maybe 300 real squads left ... is kind of silly.

Something like the medal system is better.
You get medals on a win, lose medals on a lost. But you always get some medals on a win (this, in itself rewards active squads over inactive ones, but at least you only get medals on a win, unlike KSOD which gives out skill points for losing).
You get more medals for beating a higher ranked squad, and less for beating a lower ranked squad.
Simple, but works, IMHO.

(Thanks for not directly calling me an idiot. But your assumption that there are only 300 real squads is just false. Base on our ranking of 2000 or so, there must be 2000 squads that war at least once a week. My guestimate is that there are about 3000 squads.)

And you know why people is passionate about medals ?


User avatar
Spock
Admiral
Posts: 794

Re: KSoD Squad Ranking Website - Discussion Topic

Post#220 » Fri Dec 07, 2018 12:15 pm

ChiptheFish wrote:
Spock wrote:
StrikerRebel wrote:Hi Spock, first thx for your awesome work, but one question, before last war, that we won, we had 29114 skill points, after that war we had 29094 skill points, how we lost it ?
Squad is ActiveDonatorYMW thx for response....


The 0 sigma fix caused some adjustments in the rankings. It looks like your squads sigma was adjusted to 0.01 with the rest of the squads that were stuck with a locked sigma. In the short term you might have lost 20 skill points but you will now actually be able to gain more skill points (depending on your matches) instaed of being stuck and unable to gain skill points at all.

Spock, first of all, huge thanks for the work you guys do. Big fan.
Is our squad an example of getting locked in a sigma rut? FriskyEwoks:
http://ksod.us/squad/074ce1e7-ad0c-11e5 ... df/history
Even with a string of wins, we seem to be pegged. Thanks for the comments!


Our pleasure really :)
It totally looks like you were one of the squads that was stuck with the sigma lock as you mentioned. You had some good wins there but got no skill because of it.
Now that the fix has gone out if you get a good match in the future you should have the opportunity to gain some skill points and hopefully move up in rank.
Good luck!
Image

Leader of the imperial_enclave. Flagship of the enclave brotherhood of squads.
Always looking for active lvl10 commanders. Line app required.


Return to “General Discussions”

Who is online

Users browsing this forum: Conka, EdibleKarlos, Google [Bot], Majestic-12 [Bot], Prime, RylosAlchiban and 47 guests