|
User
Soundoffs 4
Album Ratings 142
Objectivity 62%
Last Active 09-25-20 9:08 pm
Joined 11-12-11
Review Comments 2,122
|
 | Feather I still think about your awesome analytical posts
|
| Feather I still think about your awesome analytical posts
|
 | jimthefish the band Self Insert is doing some crazy stuff fusing heavy music with hip-hop and electronic music, heres the first song on their EP https://www.youtube.com/watch?v=cbIlkN-iFb0
|
 | infectionmetal Hey man I wonder if you'd like to hear some Peruvian death metal (https://www.sputnikmusic.com/bands/Infection/130265/)? I could send you a download link or a physical copy of the albums :)
|
 | neekafat Never listened to them before but perhaps its time to start (:
|
| neekafat So much stuff I need to check dropped all at once my god
|
| neekafat Good s*** hahaha you liking it?
|
 | Tyler. where did u go..........................
|
| neekafat hell yeah man!! let me know if there's anything you need from me (:
|
| neekafat Hey dude! You still on board for the Song of the Day year-end stuff?
|
 | bgillesp No problem! Just lemme know if anything needs to get changed. Also it can wait till the new year so we have a full year of stats of you want.
|
| bgillesp https://docs.google.com/spreadsheets/d/1yo9Aj7t1IZwG41xFgX9kzIugIsVckENq5RnQace1WDw/edit#gid=1865222743 Finished catching up the old months for the spreadsheet. The only issues I can force are that some song titles have parentheses (which I think your code eliminated) and one month has 2 entries for the 31st. Let me know if any of that (or anything else) needs fixing for you. Thanks again so much for doing cool stuff for us!
|
| bgillesp Gotcha. No rush at all. I was just curious
|
| bgillesp Did you post anything and I just missed it for the SOTD stuff? If not, no rush, just curious if I didn't see it
|
| bgillesp Ooh, neat. Ill have the old months' data up within the month i think
|
| bgillesp I know the stats on it are super basic, but you might would be interested in doing something with our Song of the Day thread data. We have a spreadsheet up for it in the July thread. Ill try to go back and add old months at some point but not right now.
|
 | Doctuses true I think michael jordan said that
|
 | Winesburgohio argh so many variables, the most brutal being time (consider the amount of time it's taken Yeezus to equal, at least, MBDTF or w/e) might be tricky. also i promise to be the sigur ros of interviewees
|
| Winesburgohio an idea: u know how when new albums are released their averages are invariably higher than two months after initial impressions? is there any way to chart that through Math and Statistics or are there too many variables (more users not inclined to that genre listening because of popularity, for example). anyway interested in the way ratings depreciate over time. thanks.
|
 | hal1ax ohh splendid. thank u my man
|
| hal1ax lol ya i mean the efficacy / consistency of step-wise models aside, i have to generate / interpret them correctly for my class :[ thus, i am trying to understand them in all of their convoluted glory
|
| hal1ax ya there's a lot of wiggle room for customization and tampering. although there are variant ways to go about setting them up , there's also an ambit of core concepts used to interpret them .. even tho the question i asked u, when typed into google, engendered some serious ambiguity (ie suppressor variables, mediation / moderation / better fitting model dragging prior sig.. etc.)
|
| hal1ax and when i say gaining i mean the number is getting smaller / more significant lol. poor phrasing
|
| hal1ax ayt sry to be annoying but i have a pretty basic conceptual question that's not clicking for some reason. this is a 4-step hierarchical model --- https://imgur.com/a/rQcnR --- let's take, for example, steps 2 and 3. why is, say, the significance of my selfdet coefficient (in the upper set of numbers) slightly gaining from step 2 to 3 to 4? is it because my added variables are, in combination with selfdet, making for a better fit / R-squared value? and thus it's changing their sig
|
| hal1ax sry to bug again homez.. do u happen to know anything about hierarchical / step-wise multiple regression models???
|
 | Hawks I"m still not quite sure what you're doing but yeah sure bro whatever helps you out lol.
|
| hal1ax yo mac i was thinking, would u be able to run an analysis on words which are statistically used more on sput than, say, another music site? or are u not able to scrape data from other sites? (idk how that sh1t works lol)
|
| hal1ax ya most likely. but then they would probably inquire about my acclaimed analysis of sputnikmusic m/'s research, for which i will then be at a loss
|
| hal1ax interesting.. regardless of whether it removes it or not, it sounds like it should at least give u a better idea as to which of the predictors is pulling more weight ? idk. multiple linear regression is pretty challenging. i'm in a graduate level stats class right now and am enjoying it. learning R has been a bit hellish tho
|
| hal1ax If they help to shed light on the strength of our predictors then I don?t see how it wouldn?t help us demarcate which of our predictors is explaining more of the variability in Y. It should at least help us better gauge which predictor is contributing more to our R-squared value... no?
|
| hal1ax ahhhh to get around problems pertaining to multicollinearity probably. cool thanks. ya i did a little googling myself but wasn't rly finding any concise explanations as to 'why' it is done. figured u might know
|
| hal1ax mac daddy can u explain to me wtf the point of standardized//Beta coefficients are in multiple linear regression models? ? i understand it sets the variances of the dv and predictors to 1 but like what is the point
|
 | Sniff it would be nice to be able to edit shoutbox posts so I didn't have to write a new one every time. But of course only the most active 64 users or so should be eligible for data gathering. Given that there is a way to do that
|
| Sniff and probably to keep it relevant if possible it could include users that has been active during the last year or so.
|
| Sniff band and album edits, album ratings, forum posts, lists, blog posts and review comments. With all contributing to a total sum of activity with different weights. Like reviews, lists, news articles and blog posts should def contribute more than comments and so forth. And then maybe just make some kind of average of these numbers so users can see if they are more or less active than others, and if so, in what sections of the site they are.
|
| Sniff The whole reason I started thinking about this is because I felt like some hotshots were missing from the March Madness tourney and I thought that if we had some form of system where we could see who the top users were we could even have a ranking system for March Madness. Now keep in mind I don't really know how you gather data and how much work you want to put in and all that but I'm thinking. Number of reviews * approval rate (contrib staff get 1.0), soundoffs, news articles...
|
| Sniff I see. You could probably pair it with some stats over average/mean post count or some. I have some vague detailed plan. So hmu if you feel like going for it and I can explain in more detail?
|
| Sniff Im thinking something along the lines of the most active users type of deal. Build on what you have for weighted album averages. Assigning scores for albums rated, total number of comments, reviews, blog posts etc etc.
|
| Sniff Sputnik power levels stat blog when?
|
 | Drifter Uhhhhhhhhhh symbolic, none so vile, and ghost reveries?
|
| Drifter What can I do to get you to interview me
|
 | Gnocchi https://www.sputnikmusic.com/list.php?listid=180266
|
 | butcherboy got a couple of ideas for your statniks if you fancy.. Maybe trying to determine album overlaps between big numbers of users that seem to bypass specific taste either due to how good they're perceived to be or how good they actually are (you could play them off people's pie charts).. also, how many albums people have 5'ed, and which albums seem to warrant a lot of classic ratings, both old and proven, or new..
|
| Gnocchi Hey mate, what does ?weighted user? mean? Just reading that 2016 list
|
 | Dewinged lol somehow I didn't read that part. But it's alright man, should be fun! I live in Japan so our time differences might sly a little if u wanna do it by chat, whatever suits you better mac, I'll adapt.
|
| Tyler. are you coming on to me?
|
| Tyler. https://www.sputnikmusic.com/soundoff.php?albumid=420
|
| hal1ax i started using R yesterday for a grad stats class i'm taking! ... it's scary... so very scary. have to type like 5 fckin lines of code to make a bar graph haha :[
|
 | Jots do you have a way of seeing the sput chart for best live albums for 2017?
|
 | Spacesh1p Nice new statnik post. I followed you on Github since I still want to do some Sput analysis myself once I get some time. Just been slammed with work for the last six months. Keep it up dude.
|
 | Slut yeah lol this symbol | was working too which is kinda weird
|
| Slut thats a negatory
|
| Slut Is there any way to get around the mystery diamond so I can log on????? Youre my only hope
|
 | Jom Cool, appreciate you! I sent you one more follow-up.
|
| neekafat Hmmmmmm you're probably right, just thought it might be an interesting idea!
|
| neekafat I was thinking about some kind of way to determine the frequency of swears per genre; like how often or how frequent swearing is used in hip-hop versus folk music and how one would go about finding the necessary sample size/group/idk
|
| neekafat Yo are you on the discord? Had a stats-based list idea and some thoughts I wanted to run by you
|
 | Sowing nvm scrolled thru the thread and it was talons...for some reason he put mac as the signature
|
| Sowing it says "mac" for the boxer question...that isn't you?? if not who is that lol
|
| Sowing appreciate your continued help with the HOF feature! thanks for driving it forward when I'm not focusing my efforts on it.
|
 | Toondude10 ever thought of doing a Statnik idea based on the musical taste pie charts? I think there might be something interesting you could do with that.
|
 | Divaman Never mind, thanks, it's there, it's just not where I expected it to be. (I was looking in the 3.9's based on its rating).
|
| Divaman I love the chart you did, but I totally don't understand how it works. How did the Sufjan Stevens Planetarium album not make the chart? It's rated at 3.8 with 124 votes. Not trying to bust your chops, just trying to understand.
|
 | theacademy ugh the banner is so small and weird... any idea what resolution it should be?
|
| Spacesh1p Sounds similar to me... picked it up out of necessity but I find it to be pretty enjoyable now. Seems to be increasingly common these days to need some competence with code. Only going to keep going that direction so I'm happy I got a chance to learn at work after totally missing it in school.
|
| Spacesh1p Thanks. Do you code at work or is it just a hobby?
|
| Spacesh1p Yeah I do. I'll let you know when it's up. I'm pretty slow since I'm not super experienced, just fyi.
|
| Spacesh1p Sounds like a cool idea. I have some experience with ngram but if I have some time I'll start by just trying to scrape the review text from staff/emeritus reviews.
|
| Sowing Oh boy. Sputnik doesn't know what's about to hit it!
|
| Sowing Still working on getting blog access granted from the big man. Just wanted to let you know.
|
 | TheSpaceMan i worded that last but terribly lol so like is chunk 1 all albums added between the addition of OK Comp and Master of Puppets?
|
| TheSpaceMan oh wow this is exactly what I was wondering! dont know how i missed it, thanks for the work put into that mac. just one question, are the chunks the albums added between the two you mentioned above each main genre tag?
|
| TheSpaceMan Hey dude, the other day I was wondering how sput's music tastes have changed over the years since it was first started - which got me wondering if there was a way to quantify sput's taste in music per year through rating dates. Figured I'd ask you!
|
 | 50iL Prob the most deserved staff ever ;)
|
 | evilford thanks lol. also congrats on staff :]
|
 | Flugmorph that sounds dangerous. we must prevent you from doing so!
|
| theacademy congrats on becoming the second non-reviewing staffer (;
|
| Feather Congrats on staff, this could be neat :-)
|
| Flugmorph congrats man! what are you going to do with your newfound powers?
|
 | Frippertronics oh s*** mac we're gonna need some stats on the contribs/staff asap
|
 | TVC15 Congrats on staff dude!
|
 | Arcade if Marine Le Pen wins in Fr then you're going to be demoted instantly, numbers better not fail you now
|
 | Pon Congrats on becoming Sput's chief statistician :)
|
| Jots we're considering implementing some "special duties" eventually, beyond just reviewing stuff. your talents would def be welcomed
|
 | granitenotebook if you're looking for list ideas, you could analyze what time of week, day, year, etc. get the most views
|
 | Cygnatti you should listen to more math rock lmao
|
| TVC15 It'd be cool if you butt in on the convo in my latest list a lil bit
|
 | FullOfSounds Y'know I was getting confused about my thing in Computer Science when I realized I have you to ask. You've used Python before?
|
| hal1ax haha ya i have no idea how to code. i have a couple friends that fck with java and C so maybe ill hit em up. and ya im not crazy about spss but im pretty comfortable in it at this point, and ya i fcking hate excel--microsoft office in general tbh
|
| hal1ax ya i figured ;[ thx anyway meng
|
| hal1ax im a psych major so like all we do is diff types of anova. mostly ANCOVA but sometimes repeated and two-way and sh1t
|
| hal1ax uhh it might be spss specific, idk. i just forgot which F / p value to interpret in a repeated-measures ANOVA if the equal variance test is significant and equal variances are not assumed.
|
| hal1ax i got a free cd key thru my graduate school ;p but ya i was just wondering if u used it. if u did i was gonna pick ur brain on a few things ;]
|
| hal1ax oh cool, maybe ill check that out. i have to use spss for school. it's pretty cool i guess
|
 | zaruyache "Which review here has the most comments/views?" Next project idea?
|
| TheSpaceMan a push to retackle previously established works is never a bad idea, i can get behind that!
|
| TheSpaceMan oh okay i see what youre getting at, theres a false sense of support through statistical significance when faulty data and practice plays a huge factor in that sort of verification and hypothesis testing? like this form of analysis isn't enough to completely support a claim
|
| TheSpaceMan 40 ppl is a small sample size too relative to that kind of study... esp when it was a claim that was blown up so big
|
| TheSpaceMan ohhh just saw the follow up shout, yeah thats so silly. its because a lot of that money (ie TED talks) comes from people who dont really care about the science behind the find. its sad that a lot of interest is incentivized by the money of people who are looking for a specific answer and/or dont care enough about its authenticity
|
| TheSpaceMan these life hacky effects... are you saying the push to find them is causing them to focus on anomalies stemming from bad data? or just that our focus is in weird places right now rather than statistical analysis? just wondering where you're tying that bit in (:
|
| TheSpaceMan oh wow thanks! we werent thinking in terms of p-value, with that tip they were able to get the right answer by just constructing a normal distr. and integrating for percentages... seriously mac really appreciate the help it looked way more complicated then that (which is why i hated my stat classes...). anyways that article is neat, and if i get what its saying its that finding a balance between confidence/statistical significance without needing data to be extremely deviated from the mean?
|
| TheSpaceMan H(null)= u = 1300; H(alt) = u greater than 1300; sigma = 59? sample size n=15. Consider a rejection region of x(bar) greater than 1331.26.what is the standard deviation of the test statistic? What is the probability of type 1 error?
|
| TheSpaceMan I know how to calculate a t-value and SD, so i guess i'm asking how does a test statistic have a distribution where it could have a SD? its for hypothesis testing
|
| TheSpaceMan how would one go about calculating the SD of a test statistic? trying to help a friend with stats but i took it a few years ago and i could never wrap my head around it.
|
| TVC15 What makes rating variance above and below a 1?
|
 | ConcubinaryCode So do you have an app for your true fan rating program so people can mess around with it?
|
| TVC15 What would I have to do to bring up my objectivity?
|
 | Crysis very, very cool. thanks for doing all that work haha
|
 | Titan hey rolando! please add me to your app :)
|
 | MyNameIsPencil man good work on the list, do you need my email for the personal full ranking though correct?
|
 | sixdegrees my presence on this site is vindicated thank you
|
 | SandwichBubble I'd settle for a higher quality image of his overly-enthusiastic mug, honestly~ But what do I know.
|
| SandwichBubble 5 years on Sputnikmusic, how's it feel? | http://i.imgur.com/MLwVKy5.jpg
|
 | sosabirdman the password for all of my recent alts has been 123
|
| sosabirdman i've been thinking about making a copycat alt
|
| sosabirdman nice nice. just thought it'd be kinda apt, seeing as i've lost all my enthusiasm for music, to rate everything in the database a 2.5 tbh. would be a cool way of seeing how many albums actually are in the database.
|
| sosabirdman didn't want you to write it tbh, i've got a fair handle on coding in js and python, so I'm confident enough to have a blind go lmao
|
| sosabirdman hmmm. might be viable. would it be able to run as a background process?
|
| sosabirdman you're probably not the person to ask for this, but how difficult would it be to make an algorithm at goes through and rates every album in the database for me?
|
 | Sinternet haha wow. there's another load (don't know if they are the same) that all have the same ratings like this one http://www.sputnikmusic.com/uservote.php?memberid=1072226 . Most of them are spelled in all caps. Look at the averages on Jay Rock's mixtapes and a load of them have rated it 5.
|
| bgillesp I apologize, the link you gave me sent me to a recent list, but I found the one you meant if it was the troll list. Thanks
|
| bgillesp I think I may not have explained myself well. I'm asking if you could for example find the weighted mean of all albums I have rated and compare that to my actual average rating. Or say a user has only rated Rust in Peace with a 4. Their average album rated would be a 4.6 but their average rating would be 4. I've been reading your adjusted-weight album lists, but what I'm thinking is a bit different.
|
| bgillesp Interested to hear your thoughts or if you are interested in the idea at all
|
| bgillesp Hey, I find you lists very interesting and really appreciate the work you put into them. I've been considering an idea, but I feel like with your data, you might could turn this into a result much easier. Is there a way you could calculate a user's average rating compared to the average weighted rating of every album they have rated. I feel like this could be interesting because an average rating of 3.2 for example could mean very different things if the user only rated 4 albums vs a variety
|
 | Storm In A Teacup so cool. thanks man. thanks for all the work you're doing. it's been fun learning all these stats. we gotta find a way to collectively put your lists you're doing in a sputnik blog or something.
|
| Storm In A Teacup obviously you can see what users are ranked 1st and 2nd etc...care to tell me my ranking? lol
|
| Storm In A Teacup awesome man that'd be so cool. if you can do a cutoff of 50 that'd be imo what is appropriate
|
| Storm In A Teacup well if I had a request it would be each year from 2002 to now lmao. it would show the most important albums to this sites history and it would show the evolution in sputnik's acceptance for different music. is there a way to cut out all albums that don't have a certain number of votes (somewhere around 100-300) in your methodology?
|
| Storm In A Teacup whaaaaaa that's crazy!!!!! that is so cool!! gotta look at your lists in detail more and see what's up
|
| Storm In A Teacup a better way of putting it is what you're doing should actually be used to make the site better and what I'm doing is just reccing by year and relevancy to sputniks past without people having to worry about statistics. I hadn't seen anything you'd done before or don't remember, but I hope that someone is considering applying what you've done to the site. would make it so much better and unnecessary for lists like the one I made
|
 | Spec you're the hero sputnik deserves
|
 | iloveyouall if you want to run metrics on the zine (assuming it actually happens) or anything like that, i'd be happy to accommodate.
|
| TVC15 What happened macman? Really miss your lists
|
 | PlasticRadioBand I'm able to change the rating when on the drop but it's unable to save that rating lol... You should make a YouTube tutorial or something
|
 | EvilEyez I can tell you are the man who loves numbers more than perfect words
|
 | Snake. then how are you so good at stats and s***
|
| Snake. can you do my calc 3 homework for me?
|
 | Toad ahhh k well best of luck in your future stathack endeavors
|
| Toad it might be interesting to make a list comparing rym top 100 to your new weighted top 100 - see which albums and genres are preferred here from a quantitative standpoint
|
| PlasticRadioBand Cosmic was ranked Elite on his algorithm (before committing Sputnik suicide) so I got no worries ;)
|
| PlasticRadioBand Nah this was a backup when I got banned for no reason for a second time. First time asked the mods why I got banned, I receieved an apology from a mod who didnt know why I was and unbanned me. Second time I received no answer from a mod and decided instead of waiting it out I made this one. Cosmic was born out of stress and me relieving it by being an anonymous angsty teenager lol
|
| PlasticRadioBand Wanna make a spiderchart for LPFTW if you are able to make a spiderchart or bar graph?
|
| Spec ugh reminds me of charger
|
| Spec this person looks banned
|
| Spec that not someones username is it
|
| Spec i've been shoutboxing people who i rarely interact with to interact with them
|
| Spec you might not be the prettiest girl here but beauty is only a light switch away
|
 | Kman418 i'd assume you likely know botch already but if you don't since you don't have them rated check them out like immediately, far and away the best metalcore band ever
|
|
