ManiaExchange
  • SpecialStr
  •  Search tracks for ""
  •  Search tracks built by ""
  •  Search mappacks for ""
  •  Search mappacks created by ""
  •  Search users named ""
Location: Home  Community Forums  Competitions  AuthorMania - Event Details 
AuthorMania - Event Details
Page: 6 of 6 Post Reply
24 December 2014 06:48:11  

Thanks for your explanation of how you are calculating a percent. 100 (L / AT) = (y)
AT = Total votes author received

It is not as if I do not understand basic principles of math. If I didn't, I wouldn't be where I am today :) . I needed to know what you were considering relevant data. From your response, I was able to justify your approach, which is solid, yet still doesn't provide a relation to other percentages. It is merely a grade based on information regarding the track alone and not all tracks.


haenry says:
We calculate the result for every map independently, thus the overall amount of votes on all maps is not considered.

Giving a map a percent like this is rating the map against itself and not against other maps. It is a percentage label that does not convey how many. This is similar to what the TM-Local Karma average is, except Local Karma average takes into consideration the degree in which people voted (---,--,-,+,++,+++). Still, both are just graded ratings of the tracks that do not convey, to some degree, how that score actually relates to the other authors scores.

Compromise?
By now, we must both agree to disagree or compromise, otherwise this may last quite a while! We are both right in what we are saying, yet the only difference is how each feels on how the event should be calculated. Basically, we are both trying to justify our methods to one another, when we could combine our ideas in hope of forming a better formula for next time.

I will take your advice and consider a similar method for the next time I do this event; however, it will not be 100 (L / AT) as you have suggested. I will compromise and suggest an alternative. I won't choose 100 (L / AT) because it simply grades the author without, in some way, reflecting how it relates to other tracks. It is merely a grade like TM-Karma. Therefore, I suggest a rating system for TM-Karma I will call CTR.

Community Track Rating
ALT = Author Like Total
DLT = Author Dislike Total
GT = Grand total votes for all maps
CAR = Community Approval Rating<--- "Car" has a nice ring to it :p . Technically, I could say that "each author will receive a CAR at the end of the event!", haha... would be funny

Formula
100 (ALT / GT) - 100 (ADT/GT) = CAR
Example
100 (42/650) - 100 (6/650) = 5.5384% of the Community enjoyed this map


This incorporates both of our ideas and illustrates a result based on the communities enjoyment of the map. Should we compromise with this ^^^, or agree to disagree? :s

For now, I am going to CTR(l) + ALT+ DLT and GT(f) in my CAR to go home to enjoy Christmas, have a good one!:d
Old Age Caravanner
Location: US
24 December 2014 11:33:54  
I totally agree with your compromise :)

Yet, I have to interfere with your new suggestion: According to "Distributive property" It is the same as (likes-dislikes)/ Grand Total votes, which is basically what you did you in your computation above trying to normalize everything.

Anyways, Merry Christmas :d
gado is a
Location: DE
24 December 2014 16:05:28  
Correct:done:

RD says:
I know the "normalized" results will reflect in the same manner that I have presented the AuthorMania results.

RD says:
The only difference is what it looks like.

I am glad we could agree.;)

Next time, I'll use CAR% to illustrate the scores.
Old Age Caravanner
Location: US
Page: 6 of 6 Post Reply
© ManiaExchange (mania-exchange.com, mania.exchange) 2020. • Terms and ConditionsPrivacy PolicyTop  •  Report a Problem