Some PA Statistics

1. Number of albums per year with rating equal or higher than 4.00 (the 'normal' rating not the QWR)

2. Average rating of the Top 20 albums of each year (if there are less than 20 in the Top, the average among them)

3. Evolution of the share by Sub-Genres in the Top 100 along the years

Let's have a first look, this graph shows, in red and measured in the left vertical axis, the number of albums per year with rating equal or higher than 4.00.

In green and measured on the right vertical axis, the average rating for the top 20 albums of the year

As a first comment on the data, one could say that by 1980 the number of very good albums was pretty low (only 16 equal or higher than 4.00), but the average quality was still quite high (4.14 average of the top 20 albums). The average quality would drop dramatically to 3.68 by 1983, apparently the worst year in Prog's history.

It would also seem that as from 2004 we are blessed with a lot of great albums per year (with the exception of 2010 which apparently was a weak year) although the average quality still has some trouble matching that of the golden years 1972 - 1975, with 1972 having the highest average rating, 4.30.

For those interested, here are the figures for the graph

Another fact from the data, if you want to get all the albums with rating equal or higher than 4.00, you will need to have 1083 albums.

The other thing I checked is the distribution of Sub-Genres in the Top 100 per year (or less if there are less than 100 in a given year). This graph shows it (note that years go from bottom to top). Sub-genres are represented by colours as shown in the legend at the right side.

 

Maybe some interesting facts about this one:

- Neo-Prog appears for the first time in the Top 100 in 1982 with Twelfth Night's Fact and Fiction and East's Huseg.

Here are the data figures for this graph. Green shows the first year when the sub-genre appears in the Top 100, and orange shows the peak year where the sub-genre reached its highest share in the total.

BTW, note to the Admins: it took me a bit of patience to work this out, I wonder if there's any way to access the PA's databases more easily in order to play with the data and have some interesting discussions based on data from them. One of my frequent tasks at my previous job was data analysis from relational databases so I can have fun playing with data and building up statistics.

I love stats and will look more closely at this later on.

Does this mean that the PA community broadly recognizes these albums as no less than "highly recommended addition to a anyone's prog collection", or that most of the members simply are not interested in investing any effort into listening to these albums, much less rating/reviewing them? Out of curiosity, I have sampled a few of the Top 100, <50 ratings albums and, frankly, I haven't found any "hidden gems". Some of the obscure albums I did purchase (and liked), e.g. SOT's Redwing Nest, will never make it into the Top 100.