Note: On this first day of the new year, we bring you a post from yet Another J, my college roommate, economist extraordinaire, and runner supreme. Not only has he been kind enough to provide musical hints that have turned into blog posts, but he has also been patient enough to tolerate his stories being fictionalized and told in other posts. Here, he tries to bring some rigor to the squishy practice of annual rankings. Read it. Love it. Debate it. Or something like that.
After three years of reading this blog and seeing my adventures, superior musical taste, and mediocre musical talent alluded to many times, I have finally taken the plunge and decided to contribute in a public way. The ElderJ has asked me to contribute many times and I have started many posts in my head but the topics seemed either too small or obscure to use to introduce myself to the thousands (millions? hopefully double digits?) of readers of this blog. So I decided to start with something more interesting: a music rating algorithm to rate the top albums of 2013.
I wanted to be like all the cool kids and create a list of my favorite albums of 2013. However, that is much harder than it seems, especially this year. It seemed like there were many good to great albums released in 2013, but very few truly stood out. My top few albums were obvious, but what about albums 4 through 100? I needed a way to quantify the merits of each album to accurately rank them.
The algorithm, a fancy term for math used to impress people, came to me driving to work on a random snowy day. The idea is simple: the best albums ever made consist of the best songs, the worst albums are made of the worst songs, and all of the rest fall in between. Between the best and worst albums ever exist albums that have one transcendent song and 12 other tracks of garbage, but also solid albums that do not have one “great” song but 12 “very good” songs. The algorithm provides a way to quantify that so that I could see the relative strength of each album.
To rate albums, the Music Ranking Algorithm factors in the percentage of the album that is good (i.e. the artist’s peak of the album) and how great that peak is as well as how much of an album is filler (or at a lower level than the best stuff) and how good that is. I also included a subjective “critical adjustment” to factor in how well the album met (from my perspective) expectations/hype, the introduction of new styles or elements, and how the album fits along the artist’s growth path. Sometimes releasing the “same” album twice is just what I want, but other times I expect something more or different, so I wanted to be able to account for that. When written out, the algorithm looks like this::
Rating = ((% Peak x Peak Rating)+(% Filler x Filler Rating)) x Critical Adjustment
The Peak and Filler Ratings each use a scale of 0 to 10, where:
0 = I could not finish the song because I was running to the bathroom to puke
1-2 = Will never listen to this again
3-4 = Might keep it in my iPod, but will likely skip it if it comes on.
5-6 = Will probably keep it in my iPod and may listen to it if it comes on.
7-8 = This will be in my rotation this month. Will listen or skip depending on mood.
8-9 = This will be in my rotation this year. May sometimes skip the track.
10 = This will still be in my rotation several years from now. Will almost never skip the track.
This one got a 10.
The Critical Adjustment Rating is based on a scale of 1 to 7, where 1 is low, 7 is high, and 4 is neutral, and the rest is…subjective. A score of 1 creates a critical adjustment factor of 0.97 (or -3%) and a score of 7 adds 3%, with the other possible scores falling in between. This allows a subjective tweak to the rating, but doesn’t change the overall score substantially.
With all of that out of the way, below are the top 17 albums I have listened to in 2013, fed through the algorithm. (This list consists of the albums that I have listened to more than a few times, so that I feel like I can give a solid review.)
1. Vampire Weekend – Modern Vampires of the City
2. Frightened Rabbit – Pedestrian Verse
3. Janelle Monae – The Electric Lady
This song obviously isn’t on the Electric Lady, but is just an example of how fun Janelle Monae is live.
4. Okkervil River – The Silver Gymnasium
5. Autre Ne Veut – Anxiety
6. Mike Doughty – Circles…
7. The Head and the Heart – Let’s Be Still
8. Kopecky Family Band – Kids Raising Kids
9. The Mowgli’s – Waiting for the Dawn
10. Portugal. The Man – Evil Friends
I really like this song, but the video is very creepy.
11. Little Green Cars – Absolute Zero
12. Pearl Jam – Lightning Bolt
13. The National – Trouble Will Find Me
14. Arcade Fire – Reflektor
15. HAIM – Days Are Gone
16. The Dodos – Carrier
17. Phoenix – Bankrupt!
Looking at the list, it seems like the algorithm did a pretty good job. The albums fall into an order that fits my appraisal of them and there are only a few surprises. I loved the Vampire Weekend album and thought it was one of the best I’ve heard in years. As the Elder J has mentioned, the latest Frightened Rabbit cd is not their best, but is very good. I really like Janelle Monae’s album and keep finding new favorite parts. Autre Ne Veut has some very strong songs but is scattered, so I’m really looking forward to his next album. Mike Doughty recreated some of his best Soul Coughing songs with mixed success. The Head and the Heart and Kopecky Family Band were both pleasant finds for me this year and both have a similar boy/girl vocal mix. The Mowgli’s, Portugal. The Man, and Little Green Cars are all fun and have some strong songs good for listening to with your windows down. Pearl Jam’s latest was solid, as usual. I was disappointed in the latest output from both the National and Arcade Fire. The former didn’t take enough risks, the latter took too many. I have enjoyed most of Phoenix’s work in the past, but Bankrupt was pretty terrible.
I was most surprised with the rankings of HAIM and the Dodos. HAIM’s songs are just pure sugar coated rock, but some of their songs are really really good and really really catchy. The Dodos’ latest album was one of their strongest, but apparently lacked enough truly great moments for the robot to rank it higher.
I’m looking forward to putting new albums through this test to see where they rank and to fine tune the process. Any ideas?
Thanks to the J’s for letting me share their platform. I already have some topics for future posts that should be slightly more exciting than a “best of” list. Until then, Happy New Year!