Music Technology · Music Discovery

Understanding My Musical Taste With... Data!

Plus: Otis Redding, Taylor Swift, Tame Impala, Cream, and Big Star

Understanding My Musical Taste With... Data!

After 6 issues, lots of learning, and a whole lot of music, I think it’s time for a short recap of where we stand in the JamWise tracker. As a reminder, you can find the up-to-date tracker here to see my progress and favorites. 

Here’s a snapshot from the dashboard as of this writing. I’ve reviewed 32 out of the 500 albums on the Rolling Stone list, as well as a handful of others here and there. I plan to continue mixing in a group of new albums every month to keep things fresh, and to give me perspective so I don’t get lost in the older stuff. I’m also going to start filling in my old favorite albums in the name of gathering data and building my taste profile for myself. That way, if Spotify and its competitors ever crash and burn, I’ll be ready to choose my own new jams from now on.

You may also notice I’ve started tracking the sources of my favorites. This is a long-term project - most of the albums will obviously be from the Rolling Stone list since that’s what I’ve been focused on. But later on I want to have some good data about the sources I find for new music recommendations, and I plan to have some kind of gladiator-like competition among them to see who comes out on top. The Spotify Algorithm seems to have an edge, but what you don’t see is the absolutely God-awful Maroon 5 album it made me listen to last week that scored a whopping 0 on the Taste Meter, which really pulls down the Spotify Algo’s recommendation score in the competition. I might never forgive it for that travesty.

Electrophonic Chronic by The Arcs still holds the top spot in my Taste Rating, but I have a feeling there will soon be some challengers from both the Dave’s Old Fave’s list and the Rolling Stone list. I’m also constantly re-ranking some of the albums as I get more data, so some of the numbers might end up changing. I do that basically for calibration purposes - if I find that an OK new album has a higher Brain Rating than a top-tier classic, I definitely want to update accordingly. It’s all relative, anyway, so I think it works.

An update on the Brain Rating - I’ve gone back to adjust the old “objective” ratings to match my new criteria as mentioned in the last issue. Effectively, how much do I “intellectually” like the album, vs. the Taste Rating which is how much I liked listening to it. This Brain Rating is useful because it lets me award lots of extra points to albums I find culturally significant or important even if I don’t fully vibe with them (or the other way around, *cough cough Creed*). Objectivity is overrated, but I still feel the need for another ranking besides just liking the music, because music is often about so much more than the notes and lyrics. It’s also a way of tracking what I think others might like because it’s high quality, even though it’s not going into my daily playlists.

The correlation is still there - high brain ratings more or less equal high taste ratings, and vice versa - but it’s a little less correlated than before. Plus I’m more OK with the Brain Rating not being so perfect now. I expect more adjustments to come, and I’ll revisit the Brain ratings after another few batches of reviews to see how it’s going. I’ve also been reviewing a bunch of albums that should have high brain ratings, aka the most great albums ever, so this graph will probably remain skewed high until the Rolling Stone list project is finished.

Actually, I hope it stays skewed high forever, because that means I’ve been jamming to some awesome music.

As far as my taste vs. the year of the record, there still doesn’t look like there’s a huge correlation. I’ve started noticing some trends in what I tend to like or dislike - I’m looking for ways to track them, but for now the one I’ve started tracking is a new metric. I’ve been wondering if my level of previous exposure to the album has affected my liking of the music, so I started a new metric called “Previous Exposure” (gasp, this guy’s a naming genius) and made a chart. 

This one also shows a fairly clear trend, that the albums I know the best end up with higher taste ratings, while those I have no experience with are all over the place. I think that makes perfect sense - why would I know an album super well if I didn’t like it? But it’s interesting to see it in a graph like this. I’ll also be tracking this one pretty closely, but I’ll be looking for other ways to split up the “no previous exposure” albums to see what I like and don’t like. 

Anyway, nerd time is over, on to the music.

Project B.A.E. - Best Albums Ever - 7

I guess the handy-dandy spreadsheet really wanted me to listen to Taylor Swift this week. I’ll fix it later so there aren’t repeats, but for now, ya know, I just don’t feel like it. I’d rather jam.

And guess who’s going first!

Filed under: Articles
Written by Dave Jamwise
Share