I read the latest suggestions. Before suggesting any improvements, it would be good to learn a little about the "item being improved," for example, at least understand the formulas.

Making a maximum of 150 people participate is fine.

I think it would be quite good to only count those with 10 (or more) games in the rating. This will reduce the amount of work and make the tables less cumbersome.

The workload will increase significantly.
I've written many times about removing entries. Let's have you think about why this is impossible; there are several reasons. A hint to one of the reasons: two players participate in the game; by removing one, we invalidate the result for both, so....

But in the middle of the table, there are many people with 1 or 2 matches – they played and disappeared; these should not be counted; for example, consider them in the ranking only after 5-10 games.

Rom, I understand that it’s impossible to keep track of everything – read here /topic/at/725490/

the dependence should not be on the number of games, but on time. There are a lot of players who haven't played Heroes for 3-5 years.

Let's look up the meaning of the word SEASON in the dictionary and realize that the table reflects only the results for the last 3 months. We also try to understand the difference between a seasonal rating and a permanent one.