A relation I took a look at. At first, I looked at just the average rating at every 10th percentile and logically, the average rating decreases the less popular a show is. I then decided to plot it out to see the distribution and it's interesting to see the way the ratings curve at the end, how the ratings curve towards an integer as the number of viewers decrease. The outlier laying around the 90th percentile with the average of 5.7 is funny enough, Sword Art Online for anyone interested, and the other 6 shows in the bottom left region are Utsu Musume Sayuri, Tenkuu Danzai Skelter+Heaven, Nami, Mars of Destruction, Shitcom, and Pupa. Sample size at this time is 534 users. Note that I used the average from my sample size, and not MAL's score for a show's rating. Update: Click on the image for a live interactive version. |
_NorthFeb 9, 2017 7:08 PM
| What conclusion does this lead to? Where did you take/how did you pick the samples from btw? |
 Oshii is probably the only director that loves dogs. He thinks he's a dog himself. That's right, its slime! It will dissolve your clothing slowly before my eyes! |
| Would be fun to include age and gender to see how they relate to score. |
 |
| Idk if this can be trusted unless I see some links or something tbh. |
 |
| all this says is that your sample size don't watch anime |
| I think that's a really cool graph. I always suspected that popular shows were rated higher. |
  |
| The only thing I see on that thing is a lot of dark dots :3 |
"No matter what painful things happens, even when it looks like you'll lose... when no one else in the world believes in you... when you don't even believe in yourself... I will believe in you!" |
| I chucked at SAO being an outlier. Love the divine shitty W made by Nami, Tenkuu Danzai Skelter+Heaven, Utsu Musume Sayuri, Shitcom, Mars of Destruction and Pupa. Ultimatedeath said: Click on it and you'll see white dots and the names of the corresponding anime by hovering the mouse over it.The only thing I see on that thing is a lot of dark dots :3 |
zalSep 16, 2018 5:43 AM
| Signature removed. Please follow the signature rules, as defined in the Site & Forum Guidelines. |
| How is Gintama higher than Gintama° anyway? I think you may have used some of these 1-rating trolls :D |
| Oshii is probably the only director that loves dogs. He thinks he's a dog himself. That's right, its slime! It will dissolve your clothing slowly before my eyes! |
| It's disgusting to see Shitcom alongside all those crappy titles with no purpose. People must lack the intellectual capacity to interpret the work. Unlike the others there's a point to it, a clear and poignant statement about people, society and relationships that I found to be insightful and true. The other titles down there are just literally shit that someone, somewhere excreted through his mouth by telling a friend about this great idea he had for an anime. They are PART of Shitcom and Shitcom is part of the endless cycle of human futility and dsigustingness. Truly a metacommentary on the modern, fast-paced, over-verbalizing world we live in. It should be i the 6 or 7 range with other good and poignant shorts. |
| I probably regret this post by now. |
| It's normal if you ask me. There aren't many popular anime that I didn't like to be honest, they are popular for some reason after all. Although there are also good ones that aren't popular, but the number of those can't compare with the popular ones. |
 |
| Wew.....this is a real mind fuck lol |
 |
| I'd be interested to see how much this changes, if at all, with more and more people funneled into the equation. |
"I'd take rampant lesbianism over nuclear armageddon or a supervolcano any day." ~nikiforova |
| I thought this was a run of a mill discussion, nice to see you put in some effort in this OP. As for the results, they're pretty much what I expect. Popular shows are popular for a reason meaning they're seen as good to a large amount of people. While less popular ones are on a scale of niche cult classics to hated garbage. |
| I always thought if more people view it then, it's having less rating. Just check MAL top anime vs most popular. https://myanimelist.net/topanime.php https://myanimelist.net/topanime.php?type=bypopularity Only like 12 of the top 50 popular anime in the top 50 rated anime. I remember last time OPM is number one in IMDB but as it gather more viewers, is slowly drop in ranking. _North said: Sample size at this time is 534 users. Note that I used the average from my sample size, and not MAL's score for a show's rating. Isn't using MAL's score better and more accurate since it gives a better overall picture. |
ZapredonFeb 9, 2017 9:50 PM
But it's important to remember that a movie review is subjective;it only gives you one person's opinion. http://www.classzone.com/books/lnetwork_gr08/page_build.cfm?content=analyz_media&ch=30 It doesn't matter if you like LoGH,Monster etc.If you are a jobless or college/school dropout living in your mom basement, you are still an unintelligent loser. Taste in anime does not make you a better person.If elitist don't exist, casual pleb and shit taste also don't exist. |
_North said: The average number of complete shows in my database is 512 and include all sorts of people so I'd say it's adequate for this. Faking... I mean working with statistics a lot, this sounds a little vague to me. How exactly did you pick the sample users? Apart from that, nice analysis, I really enjoyed thinking about the result. So what's your interpretation of this? I would have guessed that the more popular the show the more diverse the range of rating. But this proves me wrong. So obviously your sample has been found to be influenced by other factors than personal taste when rating. |
| Considering people rate anime differently and I believe not all of samples watched all of anime that being included in the analysis, I think it would has a "better" result if the ordinal scales have transformed into interval scores using Z-score before plotting. Assuming you pick your 534 samples randomly. Some people throw high scores so easily, some others throw low scores so easily, and some others are a bit fair in throwing high scores and low scores. That differentiation will create bias in average score if average score quantified from the raw scores. Especially for shows that has a low number of raters. Though I understand that it would be such a pain in the ass to transform the scale of 534 people into 534 interval score. |
Kurniawan_KtrFeb 9, 2017 10:31 PM
"People who don't see that anime has changed are either wearing "glasses" or watching only a certain type (and or era) of anime" "Having a low mean score doesn't necessarily mean one doesn't enjoy anime. Rating system is not a school grading system." "Elitist is people who think he is superior than others. Not necessarily ones who insulting/critisizing your favorite anime or people who enjoy a certain type of anime" "Fanboy is people who translating "your favorite anime is shit" into "you are shit". "Being a fanboy is an indication of elitism" |
n8fury said: _North said: The average number of complete shows in my database is 512 and include all sorts of people so I'd say it's adequate for this. Faking... I mean working with statistics a lot, this sounds a little vague to me. How exactly did you pick the sample users? Apart from that, nice analysis, I really enjoyed thinking about the result. So what's your interpretation of this? I would have guessed that the more popular the show the more diverse the range of rating. But this proves me wrong. So obviously your sample has been found to be influenced by other factors than personal taste when rating. I kind of assume it was the users who entered their account into the witcheffect site. Considering that's where the graph links to. So I know it was advertised in at least 1 club and there was the forum post about it a couple months back. |
_North said: Zapredon said: Isn't using MAL's score better and more accurate since it gives a better overall picture. I don't believe so if I'm not also using MAL's whole population. As well I don't believe Xinil would be happy about me scraping the average rating for 12000 shows. Zapredon said: I always thought if more people view it then, it's having less rating. Perhaps for trending shows, but if you look at IMDB's most popular and highest rated movies, 6/10 shows are on both top 10s. www.imdb.com/search/title?title_type=feature&sort=num_votes,desc (Popularity gives only currently popular movies) http://www.imdb.com/chart/top MAL's top 100 for rating isn't exactly the most stable list and is easily influenced by seasonal shows. Have you ever try doing the same stats but this time using MAL actual score(if you have time of course)? And what you mean by influenced by seasonal shows and unbalanced? You mean still airing? Maybe only completed anime should be taken into consideration since completed anime rating is more stabilized. Regardless of what IMDB suggest, MAL clearly shows that more popular shows has less rating. If you look at the bottom right, you can see the amount of users MAL has. It has total 4 million so using just 534 is just small number. https://myanimelist.net/forum/ |
ZapredonFeb 9, 2017 10:51 PM
But it's important to remember that a movie review is subjective;it only gives you one person's opinion. http://www.classzone.com/books/lnetwork_gr08/page_build.cfm?content=analyz_media&ch=30 It doesn't matter if you like LoGH,Monster etc.If you are a jobless or college/school dropout living in your mom basement, you are still an unintelligent loser. Taste in anime does not make you a better person.If elitist don't exist, casual pleb and shit taste also don't exist. |
_North said: Kurniawan_Ktr said: Though I understand that it would be such a pain in the ass to transform the scale of 534 people into interval score. Actually, I do have standardized scores as a datafield in my database for the reasons you described, and then weight and convert it to a 0-10 scale for certain scorings. I decided on raw ratings though for this since people get confused when the standardized ratings seem to make no sense to them (Though it seems to have occurred anyways). I doubt it'd make a difference in the general trend of the graph, but I will see how it goes when I get home from work. Wow. Great job man. Yeah, I do agree that it will have no much difference. I also don't think it will affect the general trend that much. I am just curious about how the extreme data will "change". |
"People who don't see that anime has changed are either wearing "glasses" or watching only a certain type (and or era) of anime" "Having a low mean score doesn't necessarily mean one doesn't enjoy anime. Rating system is not a school grading system." "Elitist is people who think he is superior than others. Not necessarily ones who insulting/critisizing your favorite anime or people who enjoy a certain type of anime" "Fanboy is people who translating "your favorite anime is shit" into "you are shit". "Being a fanboy is an indication of elitism" |
_North said: Zapredon said: Have you ever try doing the same stats but this time using MAL actual score(if you have time of course)? And what you mean by influenced by seasonal shows and unbalanced? Regardless of what IMDB suggest, MAL clearly shows that more popular shows has less rating. If you look at the bottom right, you can see the amount of users MAL has. It has total 4 million so using just 534 is just small number. https://myanimelist.net/forum/ As I said, getting MAL's score would take a couple of hours since I can only fetch so many entries before getting 429'd, as well, I'm not interested in the MAL as a whole. There are 4 million users, but the number of those still active and rate their shows reduces it considerably. Looking at sample size calculators (I am no real statistician, the most I did was discrete mathematics in university), it seems that for a population of 4 million, with a confidence level of 99% and interval of 5, I'd need a sample size of 666, which considering it's not actually 4 million, seems acceptable to me. http://www.surveysystem.com/sscalc.htm https://www.surveymonkey.com/mp/sample-size-calculator/ www.raosoft.com/samplesize.html I feel no need to reflect MAL completely, why not include in the other databases like AniDB or AnimePlanet. If you wish to skewer my results to your desire, simply add more people to my service. And yes, the most popular shows are not in the upper 8s and 9s, but they still hold higher ratings than the average. A more useful analysis would be looking at the popular shows and seeing their rating over time as their popularity/time increases, sadly I can not think of a way to find past data on this at the time. There is https://anime-stats.net/ but I believe they don't keep track of a show after it's finished airing. How and where do you get the 534 users? Did you advertise your database somewhere before or they just google it and somehow find your link? SAO is 7.8 in MAL but it's 5.86 in yours. SnK is 8.53 in MAL but it's 7.46 in yours. Does people who are not fan of SAO go to your link? Regardless of confidence interval, it seems it's heavily flawed.If you are not interested in MAL as a whole,I guess that's ok. Let just says that rating and viewership are correlate with each other,shouldn't it be the other way around where rating affect viewership. Low rating anime have low viewership because people use rating as a form of recommendation so they decided to use rating to watch anime. When an anime is rated lowly, then people would have avoided it the first place, thus the low viewership.But more importantly I don't think rating had anything to do with number of viewership. |
ZapredonFeb 10, 2017 2:37 AM
But it's important to remember that a movie review is subjective;it only gives you one person's opinion. http://www.classzone.com/books/lnetwork_gr08/page_build.cfm?content=analyz_media&ch=30 It doesn't matter if you like LoGH,Monster etc.If you are a jobless or college/school dropout living in your mom basement, you are still an unintelligent loser. Taste in anime does not make you a better person.If elitist don't exist, casual pleb and shit taste also don't exist. |
| I'm not sure what you're looking for from these questions, approx half is from some occasional advertising last year on these forums as well as friends, the rest from word of mouth. My site is not SEO'd so doubtful anyone came blindly from google. I agree with what you say about low rated shows, though I'm not sure about shorts, since people seem to never rate shorts above 7, barring a handful of real gems. Here's the graph using standardized scores.  |
| For what it's worth - although I don't know how relevant this is - when comparing the top 50 ranked against the top 50 most popular, I tend to rate the top 50 ranked far higher than the top 50 most popular. My ratings for the top 50 most popular tend to be significantly (3+) lower than their averaged weighting. |
 It's an entirely different kind of flying, altogether! It's an entirely different kind of flying. |
| Hey @Lord_Sithis . Is that Rabi Ribi or a similar game? |
Lord_Sithis said: sasalx said: What do you mean? What is similar?Hey @Lord_Sithis . Is that Rabi Ribi or a similar game? MAL, please remove the character limit. Graph looks like one of the bullet hel games. |
_North said: My theory on this is that people getting into anime tend towards the popular shows which, because they're still new, will most likely enjoy. You still have people of all different tastes watching these shows, who'll rate it anything, but a significant chunk of it are new people, which you won't find watching the less popular shows as much. I think you have a point there. I'm quite new to Anime, too, and I admit that I watched, for example, SAO just because I wanted to know the show so many people talk about. Another theory I have is that your result could (maybe) hint to how people rate Anime: so, let's assume you have two types of raters (and I know this is too generalized, but for the sake of explaining my point), the ones who are critical and use their full scale from best to worst or what not, and the ones whose average is around seven instead of five (for any reasons). If there are more raters of the last sort, then a more popular anime gets a higher rating. And this seems to be the case. And if that were true, I'd say using rating as recommendation is a bad idea. |
| Jesus....you people and your graphs.... Who honestly cares? Seriously....why do you guys care SO much about these ratings? If you like or dislike a show....that's all that matters. |
| Cowboy Bebop almost in the 7 zone? Still too high, but at least not as bad as MAL's rate. |
 |
| Popular anime are consider as mainstream or how much people watched it. But high rating anime means average mal people likes for a show There are few anime that have both element, many other are get one of them, or maybe they didn't get any So I don't get why did OP put the popularity and rating in verses mode?? |
 |
More topics from this board
» Do you find immoral anime fun to watch?LenRea - Yesterday |
26 |
by Cielord
»»
59 seconds ago |
|
» HELP ME FIND THIS ANIME!!!nakurwiamnleb - 26 minutes ago |
3 |
by Emmuhzx
»»
3 minutes ago |
|
Poll: » Do you think western animation is better than anime right now? ( 1 2 3 4 )Runasius - Feb 22, 2023 |
185 |
by SmugMiri
»»
4 minutes ago |
|
» Shoujo animeDaniellaaa - Yesterday |
26 |
by ColourWheel
»»
9 minutes ago |
|
» If you watch a series but skip the filler, can you really mark it as complete? WHat's your opinion on this?TheBlockernator - Nov 22, 2021 |
42 |
by Spunkert
»»
11 minutes ago |