| - novelty bias works only when the first episode has very good production value (then if the writing of the following episodes is bad the score will drop) or if the fans of the source material are enough (so it works like the sequel effect) - the score boost after the last episode only happens if the ending is very well liked (Odd Taxi, Steins;Gate); the score remains more or less the same more often (a little boost quickly nullified by score degradation) - recency bias and score degradation are both caused by the fact that die-hard fans of a specific genre will watch a related anime as soon as possible, whereas those who care less will take their time other ideas: - controversy bias. The more explicit an anime is in terms of violence and sexuality, the lower it will be rated - popularity bias. The more an anime is popular, the more it will be watched by newcomers, which have lower standards, thus inflating its score - "status of classic" bias. The score of some old anime is increasing instead of degrading (Monster, Nana, Banana Fish, Serial Experiments Lain to name a few). I think it happens because they achieved the status of "classic", so new watchers are seeing them more positively - bandwagon effect. When a new season comes out and it's rated much higher than s1, the score of the latter will raise as well (Fruits Basket, Kono Oto Tomare!, Dragon Maid). I'm not sure if it's just the influx of new watchers, or there really are so many user fixated on giving the same score to every season of a franchise. - Netflix jail effect. Beastars score falling from 8.42 to below 8.00 in a little more than a year is unprecedented. My own explanation is that those who watched it on Netflix liked it much less than those who did it on illegal streaming sites when it was airing. Why? It is reasonable to think that those who'll bother using obscure ways to watch anime are tendentially more expert watchers and thus less outraged by the furry content. |
deg said: - novelty bias - first episode usually have high scores - usually the ending episode will have a good boost in the overall score of the anime - sequel effect - sequels usually have higher scores since the haters stop watching and only fans remain watching the sequels - recency bias - new anime tend to have higher scores - old anime tend to have decreasing scores over the years - there are more binge watchers than ongoing weekly watchers - currently with 12 million members once an anime reach 180k members then its part of top 1000 popular anime out of the current 19K anime in the database so can you add more? and thoughts? - novelty bias - first episode usually have high scores (agree) - usually the ending episode will have a good boost in the overall score of the anime (agree) - sequel effect - sequels usually have higher scores since the haters stop watching and only fans remain watching the sequels (agree) - recency bias - new anime tend to have higher scores (agree) - old anime tend to have decreasing scores over the years (agree) - there are more binge watchers than ongoing weekly watchers (can i get data from it ? - currently with 12 million members once an anime reach 180k members then its part of top 1000 popular anime out of the current 19K anime in the database (agree) |
 |
deg said: Since you seem to believe art is subjective, you could rename these to "novelty factor", since bias seems to imply a departure from objective fact. Instead, what you mean to say is that novelty, recency, etc., affect the subjective impression. Not that there is some intrinsic value to art, and novelty, recency, etc., throw our judgment off.- novelty bias If you do mean the latter, then you could not possibly think art is subjective. These are contradictory views. |
| My subjective reviews: katsureview.wordpress.com THE CHAT CLUB. |
deg said: @Nirinbo controversy bias and Netflix Jail effect are great additions im not sure about popularity bias though, since the more the anime becomes popular the lower its scores get overtime due to 1/10 scores having more weight than 10/10 scores Yours is a valid hypothesis, but I'm not really sure. On the contrary, I'd say that the 1/10 votes hurt less popular series even more precisely because they aren't watched by newcomers who rate everything 9 or 10 (for example, Aria the Origination: 8.99 on RedditAnimeList, 8.50 on MAL). |
| - all anime that have a score close to 9 while airing wiill be downvoted by fmab fans |
    |
Nirinbo said: - popularity bias. The more an anime is popular, the more it will be watched by newcomers, which have lower standards, thus inflating its score I don't think this bias gets enough acknowledgement. - Netflix jail effect. Beastars score falling from 8.42 to below 8.00 in a little more than a year is unprecedented. The opposite also holds true, due to popularity bias. Movie scores often go up after they hit streaming services. |
| その目だれの目？ |
Lucifrost said: - Netflix jail effect. Beastars score falling from 8.42 to below 8.00 in a little more than a year is unprecedented. The opposite also holds true, due to popularity bias. Movie scores often go up after they hit streaming services. This is indeed true in most cases, due to popularity bias. I called it "Netflix jail effect", but actually it's something that specifically addresses Beastars. I think it's due to its controversial nature coming out only towards the end of the season, whereas Dorohedoro (another victim of Netflix jail and also possibly controversial) would be instantly skipped by those who can't stand violence. I just checked on Wayback Machine: Beastars score only fell by 0.02 two months after it finished airing in Japan, when it still was in Netflix Jail. Then it was released worldwide and lost a whopping 0.20 in three months. it can't be a coincidence. |
Nirinbo said: I just checked on Wayback Machine: Beastars score only fell by 0.02 two months after it finished airing in Japan, when it still was in Netflix Jail. Then it was released worldwide and lost a whopping 0.20 in three months. it can't be a coincidence. I think that without Netflix jail the score would simply have dropped more quickly. Erased was in the top 10 while it was airing, but scores like those never last after the hype dies down. |
| その目だれの目？ |
| Unpopular anime generally have scores that are a closer indication of their quality since the only people who would bother to watch them are those who actively seek out anime apart from the norm. The last episode boost is due to those dedicated to sticking through the show are the ones who liked it. |
 Latest reviews: Summer Pockets and Train to the the End of the World |
| MAL rating does not define whether an anime it is good or not. |
Lucifrost said: Nirinbo said: I just checked on Wayback Machine: Beastars score only fell by 0.02 two months after it finished airing in Japan, when it still was in Netflix Jail. Then it was released worldwide and lost a whopping 0.20 in three months. it can't be a coincidence. I think that without Netflix jail the score would simply have dropped more quickly. Erased was in the top 10 while it was airing, but scores like those never last after the hype dies down. Erased peaked at 9.14 while it was still airing, what matters is the score right after the final episode. score right after it finished airing -> 2 months later -> 3 more months later Erased: 8.76 -> 8.75 -> 8.69 Beastars: 8.44 -> 8.42 -> 8.23 |
| Come on you people. It's 2022. Paying attention to MAL's scores/system/lists etc is extremely pointless. You're just wasting your time. One number doesn't necessarily justify a title's quality anyway |
PokitaruJan 16, 2022 9:11 AM
 |
More topics from this board
» What percentage of your waifu's body is kissable?Catalano - Oct 12 |
24 |
by peodsgns
»»
5 minutes ago |
|
» Has using MAL made you enjoy anime less?Dragevard - Oct 8 |
42 |
by Stanis150
»»
7 minutes ago |
|
Poll: » Has BL animanga been getting better?Auron - Oct 12 |
20 |
by Cherie_758
»»
22 minutes ago |
|
» Anime characters that used to be older than you, now you're older than themComeInReiAsuka - Oct 12 |
21 |
by ApfelMyName
»»
27 minutes ago |
|
» 🍷 AD Summer 2025 Best Girl Contest ( 1 2 3 4 5 ... Last Page )Shizuna - Sep 28 |
431 |
by fi_st
»»
30 minutes ago |