I actually agree with you on this, Henrik. 4000-5000 albums= nearly a top 100 for each year, and probably more than 100 for several years.
You have to ask yourself if there is a point where an album's acclaim is so low, that it's just a statistical anomaly. I always figured (ideally) if every album on AM was reissued and reevaluated, most critics would give said albums roughly 8/10 or higher (even if initial meta-critic reaction was mixed.)
The second you go into big numbers, you start bumping into more albums that aren't necessarily acclaimed as they are just a personal favorite of one or two publications. I think 3,000 is a good place for quality control purposes right now.
I agree when you get to 4000-5000 the statistical variability gets too big to be relevant.
Something that I'd be interested in is if there's a special category of BU for albums with top-heavy praise. Some kind of 'Divisively Acclaimed' category.
If you ask a hundred different people to name the best album of all time, the chances are that you will not get a unanimous answer.
This web site's purpose is to achieve a means of establishing the greatest music albums of all time by using legitimate critical sources and calculating a ranking based on the rank and number of times an album has appeared in a 'greatest album' chart and then ranking the albums according to their aggregate performance.
===============