Forum - View topicThe Best and Worst of the Season So Far: Week of August 5-11
Goto page 1, 2, 3 Next Note: this is the discussion thread for this article |
Author | Message | |||
---|---|---|---|---|
meiam
Posts: 3471 |
|
|||
It'll be interesting to see how the ranking next week will be affected now that people know the score and series are compared every week (I guess, is this going to be a weekly feature?).
I would guess this would create a "min-maxing" effect were we see more maximum score for series people really like and min scoring for series people hate watch or don't watch. I would love to see some form of score distribution. |
||||
killjoy_the
Posts: 2501 |
|
|||
That'd be so silly and I can totally see people doing it As for the graph, just look at Gatchaman go. They REALLY don't like standing in one place, huh. |
||||
myskaros
Posts: 604 Location: J-Novel Club |
|
|||
Prison School is rising from the ashes, lol.
|
||||
Suncraft
Posts: 44 |
|
|||
Don't think this should be every week. Maybe every 3 weeks, or Start-mid-end of a season. Changes might mean something.
Going every week you might as well have a link at the top of ANN to a live result sheet. edit: Really wish people had gotten Symphogear, Prisma Illya, and Himouto! Umaru-chan up on the reviews to see their positions. Each of them have dedicated followings by now, and Symphogear is really rocking the stage in Japan pre-orders at #1. Hype at maximum everywhere but here it seems. Your poll to decide which shows got reviews probably should of been open longer. Or, maybe, like Shimoneta, had delayed entries. |
||||
DK1105
Posts: 16 Location: Detroit, MI |
|
|||
My Love Story has been great and Wagnaria is always fun but really don't see how people like Charlotte. I only saw the first two episodes and found it almost unwatchable. Maybe I missed something. My favorite show has been Baby Step but I understand why that's not on the top of the list.
|
||||
Marzan
Posts: 521 |
|
|||
I find ANN's dislike of Arslan absolutely baffling. It's far from great, has awful 3D animation for it's battle scenes but probably better than half of the stuff in front of it.
Is it the lack of female characters? |
||||
meiam
Posts: 3471 |
|
|||
I doubt it has anything to do with the quality, it's most likely because it's a "shocking" show, so lots of people turned in to watch the first episode or two but gave it terrible rating while the smaller number of people who are genuinely interested in the show consistently rate it high. So after the first few episode the people who rate it low stopped watching and rating while the people who rate it high stick around. I would guess the number of vote took an equally dramatic shift, only going down in that case. |
||||
myskaros
Posts: 604 Location: J-Novel Club |
|
|||
It's definitely not a show I'm watching, I just thought, in the graph of the cumulative ranking, going from 20 to 7 was amusing |
||||
Via_01
Posts: 551 |
|
|||
The accuracy of these charts still irks me. There are probably many people (like me) who simply ignore or forget to rate the episode they just watched, and there's also the fact that this is taking the average rating of voters, but without having the same number of voters for each series.
For argument's sake, let's say that Gintama has 20 people voting in every week, while Non Non Biyori has 200. Even if they get more or less the same amount of voters, one of them clearly has a much wider audience than the other. I don't know, I simply can't see it as a good representation of how people think of these shows. |
||||
Angel M Cazares
Posts: 5523 Location: Iscandar |
|
|||
I hope people are mature enough and that they don't intentionally skew their episode ratings. I really like this ranking feature, and I would hate to see it ruined. |
||||
killjoy_the
Posts: 2501 |
|
|||
I think the best way to deal with these worries is if at the end of the season (or maybe some other arbitrary point?) they release a feature just like the current Top 10 page is done (with different kinds of statistics, rankings based on different averages, etc). Though in the previous thread I remember Zac saying they wouldn't (or at least didn't have plans so far) release all the numbers. |
||||
Hameyadea
Posts: 3679 |
|
|||
Gintama retaining its crown, while Chaos Dragon: Sekiryū Sen'eki scraps the bottom.
As others have asked, does the graph shows the data as-is, or is it calculated and adjusted for the different numbers of votes each show gets? For example, if 300 users rated this week's Gintama episode (#284), 120 users rated Gate: Jieitai Kano Chi nite, Kaku Tatakaeri's 7th episode, and 650 users rated episode #424 of Naruto Shippūden, how does it reflected in the graph, if at all? |
||||
Dan42
Chief Encyclopedist
Posts: 3794 Location: Montreal |
|
|||
If the ranking does not match your opinion, that's entirely normal. So don't be upset about that. You are not supposed to agree with the ranking. It's just an average of opinions, published for informative purposes.
I'm not entirely sure I should talk about how the ranking is calculated, but since so many people insist, oh well. Basically, all the shows are compared one-to-one. If, among all the people who rated both A and B, 60% prefer A and 35% prefer B (with 5% equal), that means A "wins" over B and will be higher in the ranking. Usually. It's possible to have a situation like A wins over B, B wins over C, C wins over A. In that case the order is decided based on the greatest win margin. See Schulze method for the full details. |
||||
Hyperdrve
Posts: 276 |
|
|||
It would be nice to know how many people voted for a particular episode along with its ranking.
|
||||
Zac
ANN Executive Editor
Posts: 7912 Location: Anime News Network Technodrome |
|
|||
We aren't releasing vote totals. Further calls for that information will be ignored - sorry, it's internal traffic data. Not for the public.
This is just supposed to be fun info, not a scientific ranking of objective best to worst. Nothing more than some fun data we get from our episode reviews. |
||||
All times are GMT - 5 Hours |
||
|
Powered by phpBB © 2001, 2005 phpBB Group