Innate Imbalance In Tethra's Doom Scoring

[Zerrien]

Does DE acknowledge that a 300 Player Clan can't possibly beat a 1,000 Player clan, given that a 1,000 Player Clan only needs to perform at... 31% efficiency to outright beat a 100% efficiency 301 Player Clan?

 

I've covered this in feedback with my own ideas on how to fix it, but what does the community think about this blatant discrepancy in scoring?

 

Everyone gets their free Wraith item, but Golden Trophies have always been high sought after. Does this deserve to be discussed, or are we just going to accept it as unchangeable?

 

Edit: Fixed 300 to 301, just for clarity I'm talking specifically about Moon Clans.

Edited March 12, 2014 by Zerrien

[Babellon]

I was under the impression that each clan in its own bracket (ghost,shadow,moon etc) gets a top 3 trophies. a ghost clan with 10 members obviously cannot compete with a moon clan with 1k members. 

[Halser]

it's been pretty obviously imbalaced since the (third?) scoring change in the Cicero Crisis. The most obvious way to fix it would be to score via averages while making sure that only clans with 300+ members are actually moon clans(and so on) instead of basing it via what Hall they have in their dojo.

[S.Fayden]

Either having the scoring based on clan averages or total accumulated points would work just fine for this event. The current setup is ridiculous as it makes it literally impossible for some clans to win, but sadly I doubt they're going to change it.

[Zerrien]

I was under the impression that each clan in its own bracket (ghost,shadow,moon etc) gets a top 3 trophies. a ghost clan with 10 members obviously cannot compete with a moon clan with 1k members. 

 

But how does a Moon Clan with 301 members compete with a Moon Clan with 1,000 members?

[The_Herrick]

I was under the impression that each clan in its own bracket (ghost,shadow,moon etc) gets a top 3 trophies. a ghost clan with 10 members obviously cannot compete with a moon clan with 1k members. 

 

Moon clans are between 300 to 1000 members.

[Babellon]

But how does a Moon Clan with 301 members compete with a Moon Clan with 1,000 members?

hm, valid point. Not sure how the scores are tallied or have been in the past. this being my first event and all. How do they normaly do it?

[S.Fayden]

I was under the impression that each clan in its own bracket (ghost,shadow,moon etc) gets a top 3 trophies. a ghost clan with 10 members obviously cannot compete with a moon clan with 1k members. 

That's true, but the concern is the imbalance within those brackets. A moon clan can have anywhere between 300 - 1000 members. That's a pretty big range there. Normally it wouldn't be a problem, but since this event's scoring is based on the sum of each member's best run with a limit to how good that run can be, it  makes it impossible for certain clans to get anywhere near the other clans score after the majority of their members did a high scoring run.

[Zerrien]

hm, valid point. Not sure how the scores are tallied or have been in the past. this being my first event and all. How do they normaly do it?

 

For some events, they had a floating point representative of score. E.g. 'how fast players can complete a test,' this helped because the absolute best of the best were able to demonstrate their scores.

 

For others, it was a cumulative score based on how many enemies killed. This shows dedication, allowing top tier clans to farm their way to #1, simply to prove that they are the best.

 

Cicero... got complicated, they averaged the best scores of each player, but constantly shifted the rules. There was the threat that a 1-person Moon-rank clan could have the absolute best score, because it didn't need to be averaged against another clan who needed to deal with a spread (even of 1) would result in loss. Rules were changed quite a few times, and many people were quite upset. It was NOT a good way to handle it.

[Nilgreed]

This event was (one of) the best so far.

 

I've covered this in feedback with my own ideas on how to fix it, but what does the community think about this blatant discrepancy in scoring?

 

 

My opinion? I dont care about this.

I enjoyed the event - helped friends achieve their desired score and didnt get frustrated because other clans are above me on some fictional list.

[Babellon]

oh ok....sounds like they've altered their events a few times. oh well, I only scored 1567 or some such. Im not a math wiz i wouldn't begin to know how to make something like this more fair score-wise.

[Zerrien]

This event was (one of) the best so far.

 

 

My opinion? I dont care about this.

I enjoyed the event - helped friends achieve their desired score and didnt get frustrated because other clans are above me on some fictional list.

 

I appreciate your input, really. But I hope you'll sometime take a moment to look at the other clans out there who are actively striving to prove themselves on this 'fictional' list. Just as you wanted to get the appropriate score to get the Wraith, or Quantum Badge, other clans would like to have a fair footing to get a Gold Statue.

[Phatose]

That's true, but the concern is the imbalance within those brackets. A moon clan can have anywhere between 300 - 1000 members. That's a pretty big range there. Normally it wouldn't be a problem, but since this event's scoring is based on the sum of each member's best run with a limit to how good that run can be, it  makes it impossible for certain clans to get anywhere near the other clans score after the majority of their members did a high scoring run.

 

There's always going to be an issue of that kind no matter what scoring method they use.  When it's based on total of the members, larger clans have an advantage.  When it's based on average, smaller clans have an advantage - this is most visible at the lower tiers.  A 1-man shadow clan just needs 1 great run, while a 5-man shadow clan needs all 5 members to have great runs.  Not to mention that the temptation for those 5 man clans to kick the 4 lowest scoring members would be something to behold.

 

Someone is always going to have a leg up because of the rule structure.  There's nothing to be done for it, unfortunately.

[Letter13]

There are a lot of ways that scoring could have been handled better. It might have even been prudent for DE to ask the community for suggestions on scoring systems to use (i.e. clan averages, top 10% or 25% of clan members get counted, etc).

[Zerrien]

oh ok....sounds like they've altered their events a few times. oh well, I only scored 1567 or some such. Im not a math wiz i wouldn't begin to know how to make something like this more fair score-wise.

 

Honestly, it's a VERY hard problem to solve. It needs to mix elements of an average (to bring down a higher-popularity clan's score,) while also having some weighted elements to boost the scores of people who absolutely are the most skilled in a clan.

 

I'm not proposing a solution over night, but there have been tried and true events that were fair, and balanced.

[S.Fayden]

There's always going to be an issue of that kind no matter what scoring method they use.  When it's based on total of the members, larger clans have an advantage.  When it's based on average, smaller clans have an advantage - this is most visible at the lower tiers.  A 1-man shadow clan just needs 1 great run, while a 5-man shadow clan needs all 5 members to have great runs.  Not to mention that the temptation for those 5 man clans to kick the 4 lowest scoring members would be something to behold.

 

Someone is always going to have a leg up because of the rule structure.  There's nothing to be done for it, unfortunately.

The concern wasn't whether or not anyone had an advantage. The concern was that it is literally impossible for smaller clans to win no matter how well they did.

[Phatose]

I'm pretty sure that's actually a concern over whether or not someone has an advantage.  The only difference is one of magnitude.

[Psychosist]

What if we could use the current "score = sum of points" system that DE has, but hard cap the amount of players' scores that can be taken into calculation?

For example, a Moon Clan with 1000 members will only count the highest 300 or so scores among that particular clan.

This way, a 400-man Moon Clan will still be able to compete with a 1000-man Moon Clan, as only the top 300 scores will be counted in the "sum of clan points".

Ghost Clan (10 max) = Counts top five scores

Shadow (30 max) = Counts top ten scores

Storm (100 max)= Counts top thirty scores

Mountain (300 max) = Counts top one hundred scores

Moon (1000 max) = Counts top three hundred scores

Edited March 12, 2014 by Psychosist

[Zerrien]

What if we could use the current "score = sum of points" system that DE has, but hard cap the amount of players' scores that can be taken into calculation?

 

For example, a Moon Clan with 1000 members will only count the highest 300 or so scores among that particular clan.

 

This way, a 400-man Moon Clan will still be able to compete with a 1000-man Moon Clan, as only the top 300 scores will be counted in the "sum of clan points".

 

I thought about this, but, it should be fair across every clan in the group, and this is also unfair on it's own.

 

It's more likely that there are 300 excellent members of a 1,000 member clan, than there are 300 excellent members in a 400 member clan. Also ignoring the fact that a 301 member clan has next to opportunity to win against a 1,000 member clan, as every single member must be perfect, while only 3/10ths of 1,000 member clan need to be perfect.

 

A quick edit: And this is why scores should be somewhat... weighted. A 1,000 clan full of people who score perfectly should beat a 300 player clan, who scores perfectly.

Edited March 12, 2014 by Zerrien

[Phatose]

A little better, perhaps, but it still favors larger clans, as Zerrien noted.  It would also reduce participation drive - it's hard to motivate anyone to play if there's a good chance their score simply won't count, and in a 1000 man score, that means 700 people simply aren't going to count. 

[(PSN)CrimsonShinku]

we could just have any clan above mountain size go by average score, while any clan below that size goes by sum of points.

that way a 1-man clan isn't going to beat a moon clan. or any clan at all.

 

the problem is with the distribution of clans, they're far too wide apart. comparing a storm clan of 30 people and one of 100 people isn't fair. but it's still a small amount of people, in the case of mountain and above, there should be enough participants to at least make the clan (ex: a mountain clan in which less than 100 participate, why is this a mountain clan?)

 

but then if you have two moon clans, one with 300 and one with 1000, how do you determine the winner? if it's by sum of points then the 1000 clan is going to win by sheer spam, even if the 300-man clan is more close-knit and gets, on average, far better scores.

 

if we do pure averages however even if it's of those who only participated, the 1000 man clan is at a disadvantage because it's far easier to organize 300 than 1000. however this method seems a lot more fair on these scales because then the big clan wasn't acting like a clan and helping people during the event.

 

but then we downsize to storm clan, and is it fair for 30 people which is a veeery easy number to organize, to be able to beat one hundred, which is the number where things start getting out of control with everyone having different lives and such? i can't quite agree.

 

best thing i can propose is to have specific rules for each tier of clans.

ex:

ghost clans: all points added to figure the winner

shadow: all points added to figure the winner

storm: ???

mountain: averaged from the participants, at least 90 members must participate

moon: averaged from the participants, at least 270 members must participate.

 

the 90/270 members are because why do you have a clan of that size, if you can't even fit the minimum?

 

one rule i feel the game lacks right now for the events, is not allowing newly-added members during the event to count towards the total. all this does is create an atmosphere of "i'll add random people to bloat my score, then i'll kick them once my clan wins!" events should only consider the clan as it was at the start of the event.

[KONAir]

Clan average works even simplier, only the players that participated in the event are counted towards average. It was used before but for the last two events there is a childish reason we have this awkward scoring system. It will be clear soon enough and you can have a laugh at it.

[Tulzscha]

Moved to the proper section.