SOURCE

Usage Rate, Appearance Rate, Full Star Rate: https://akashadata.com/
Average Constellation, Signature Weapon Rate: 提瓦特小助手
Featured Image: Weibo@__QuAn [follow him/her on Weibo]
*This artwork is either permitted or have obtained permission from the artist at the time of article publication. If future intentions change, please contact me and I will remove the image.

DEFINATION

Usage Rate = Number of Full Stars Clearer Who Used This Character / Number of Full Stars Clearer Who Own This Character
Appearance Rate = Number of Full Stars Clearer Who Used This Character / Number of Full Stars Clearer
Full Star Rate = Number of Full Stars Clearer Who Used This Character / Number of Player Who Used This Character

MAD: Mean Absolute Deviation
SUM = Average Usage Rate + Appearance Rate + (Full Star Rate)/2
CM+1 = Constellation-Mean + 1
SW = Signature Weapon Rate
PULLS = (AC+1)×94 + SW×53
*The expected pulls to get a 5✩ Limited Character is 94, while for a specific 5✩ Limited Weapon it is 107.
GAC: Good and Cheap
GAC = (SUM-MAD) / PULLS

NOTE

1. 1. Only a portion of the 4✩ and Standard 5✩ were counted.
   2. The average equipment of the characters for this data source is relatively high.
   3. The Appearance Rate takes into account the Character’s Owner Rate.
   4. The Full Star Rate only makes sense when the appearances is sufficient. In addition, Carry has the greatest impact on this rate.
   5. MAD reflects the degree of data dispersion. It may represent the versatility of characters with low Constellation, or whether they are easily targeted by the environment.
   6. The bias in Usage Rate is often high. I included Full Star Rate in SUM.
   7. If a character’s Usage Rate and Appearance Rate are both low, then SUM largely depends on the Full Star Rate. There is some problem here.
      For example, some people will arrange a Support in a strong trio team, but rarely do so for a Carry. This leads to an inflated Full Star Rate for support characters.
   8. When calculating SUM, I divided the Full Star Rate by 2; it’s a number I just came up with off the top of my head so don’t ask me why.
   9. GAC may reflect how useful a character is under low Constellation.

    

   1. 只统计了一部分的四星角色和五星常驻。
   2. 数据来源的平均练度高于全部玩家的平均值。
   3. 登场率将持有率纳入了考量。
   4. 满星率只有在出场数足够多的情况下才有意义。另外，主C角色对满星率的影响是最大的。
   5. MAD体现了数据的离散程度。在一定程度上能代表角色在低命下的泛用性或是否容易被环境针对。
   6. 使用率的偏见程度往往很高。在SUM中我将满星率也加入进来。
   7. 如果角色的使用率和登场率都很低，那么SUM的值很大程度上取决于满星率。这里显然有不合理的地方。
      例如，会有人将辅助角色安排在强大的三人组队伍里，但是很少对主C角色这么做。这导致辅助角色满星率虚高。
   8. 在计算SUM时我将满星率/2，问就是随便设置的。
   9. GAC也许可以体现角色低命下的好用程度。

   
   GAC-15