Other notes: I am working on connecting 11 AM trace to the 12 PM trace so that there isn’t a permanent gulf in the 11>12 slice.
I am usually listening to Spotify when I’m working.
You must be in a good mood on Friday mornings and socialize more?
i don’t know what they mean, but they are very beautiful!
From a clarity-of-presentation standpoint, your graphic would very likely be much more readable with a single twenty-four hour circle, rather than two similarly-coloured overlapping series.
From a data-clarity standpoint, you’d be better off with bars/segments/’pie slices’ for each hour, where the width of the slice represents the time(s) being reported, rather than a line series. (Is the dot at 9 o’clock for the period from 9-10, or from 8-9, or from 8:30-9:30, or just a highlighted point that’s part of a more granular data set?)
There’s something wonky about the songs-per-hour scale. It looks like you’re listening to more than 200 songs per hour on Tuesdays around nine, which implies that the average listen is less than twenty seconds…?
Even assuming that you’re listening for the entire 60 minutes, you’d expect to hear no more than 15 or 20 songs per hour–unless you’re skipping ninety-plus percent of tracks.
Overlaying AM and PM was an absolutely harebrained move
![[OC] Radar Chart of My Spotify Listening Pattern On Each Day of Week in 2024](https://www.europesays.com/wp-content/uploads/2024/07/0mbmexdcuhfd1-900x1024.png)
5 comments
Source: I make a request to the Spotify API ‘[user_currently_playing](https://developer.spotify.com/documentation/web-api/reference/get-the-users-currently-playing-track)’ endpoint. I request from this endpoint every minute and save it to a database. With SQLAlchemy I returned the results of every song I listened to in 2024. I used Plotly Graph Objects following the [documentation ](https://plotly.com/python/polar-chart/)to learn how to make an overlapping radar chart, and then used SubPlots to make a grid to fill in these Polar Charts.
Tool used: Python, Plotly
Other notes: I am working on connecting 11 AM trace to the 12 PM trace so that there isn’t a permanent gulf in the 11>12 slice.
I am usually listening to Spotify when I’m working.
You must be in a good mood on Friday mornings and socialize more?
i don’t know what they mean, but they are very beautiful!
From a clarity-of-presentation standpoint, your graphic would very likely be much more readable with a single twenty-four hour circle, rather than two similarly-coloured overlapping series.
From a data-clarity standpoint, you’d be better off with bars/segments/’pie slices’ for each hour, where the width of the slice represents the time(s) being reported, rather than a line series. (Is the dot at 9 o’clock for the period from 9-10, or from 8-9, or from 8:30-9:30, or just a highlighted point that’s part of a more granular data set?)
There’s something wonky about the songs-per-hour scale. It looks like you’re listening to more than 200 songs per hour on Tuesdays around nine, which implies that the average listen is less than twenty seconds…?
Even assuming that you’re listening for the entire 60 minutes, you’d expect to hear no more than 15 or 20 songs per hour–unless you’re skipping ninety-plus percent of tracks.
Overlaying AM and PM was an absolutely harebrained move