Official Albums Chart on 11/6/2021

According to my calculations above, it's the album with the biggest chart impact by some distance. 72 places better than any other and counting.

Not that many compilation albums feature new artists because the people that buy them tend to be 40+. The compilation album has been taken over by the playlist on streaming services which skew younger.

Probably more accurately, newer artists simply don't have a long enough career or enough big hits to be able to release a compilation. In the physical era, that would be no earlier than the 3rd album (usuallu at least 4th album) with around 4 singles from each. At one album a year, that would be a 5 year career including the year before debut release to record and promote the new release.

Albums or the modern equivalent (playlists / mixtapes etc) tend to be further apart these days, 2 years often. On that basis, a modern artist would need at least a 10-year long career to consider a greatest hits. As you say, with tracks released after 2007, they are amalgamated into streaming playlists almost from new, rather making a compilation redundant for artists who began their chart career after 2005.

Sure, the thing is there never used to be this many hits album filling up the charts, let alone the top 20. A lack of interest in modern music perhaps..

There's plenty of album entries below number 53. If you have the full list - numbers 101 and below - there are a lot of new entries further down.

Take the tin foil hat off!

You know that the numbers from 101 are unimportant. Few people get to see them. This is about manipulating the chart to make it look more user friendly, but it's being done clumsily. Given the small numbers of sales involved, it just doesn't sit right.

Streaming services have the power to push certain tracks and albums and companies may pay for that to happen - eg in the week before the Brits when the old timers were pushed off streaming playlists in favour of the Brit artists - but the OCC just presents the figures they're given in a chart format. That's all they do.

You know that right?

If there are 10 records on chart and only one gets promoted, and it coincides with a big event like the Brits, then it will be the only one to gain sales. Therefore climb above the other 9, makiing one climber and 9 fallers assuming the sales of those other 9 are similar to a previous week's sales. It's a simple rank-order operation, and heavily biased due to the event and it's record label promoting it rather the UNbiased probability model you are using.