An analysis of BKRs simply indicates grades get inflated with activity For example if you look at the most popular and least popular games on the site (based on the current number of games being played) you get the following:

(a) No of established grades
(b) Top Grade
(c) Number of grades over 2000
(d) % over 2000

Most Popular

1 Ludo
(a) 327
(b) 2350
(c) 222
(d) 68%

2 Backgammon
(a) 670
(b) 2634
(c) 571
(d) 85%

3 Chess
(a) 513
(b) 2717
(c) 162
(d) 32%

Least Popular

1 Alquerque
(a) 19
(b) 2030
(c) 2
(d) 11%

2 Jarmo
(a) 25
(b) 1713
(c)0
(d) 0%

3One Way checkers
(a) 11
(b)1873
(c) 0
(d) 0%

You have to be a very strong player (or have a very strong computer) in an unpopular game to get a grade of over 2000. You would need to win a lot of games and have a perfect record. Compare that with backgammon where 85% of grades are over 2000. You are almost guaranteed to have a grade of over 2000 with a half decent record because everyone you play against has a grade of over 2000.
Do the statistics indicate there are lots of chess computers being used? Not to me. Yes the top grade is higher than Ludo and Backgammon but the number of grades over 2000 is a much smaller percentage of the total population. If a high grade indicates cheating then there is a lot more going on in Backgammon and Ludo, which I would have thought was impossible.
The conclusion I draw is that
1. Grades across games are not comparable.
2. Grades are inflationary
3. There will be no correlation between a BKR and a FIDE grade.
Why is the system inflationary? I don’t know but I guess it has something to do with good players staying on in the system and weak players not. For example if 2 players with a 1500 grade play, the winner gets say 1600 and the loser gets 1400. The loser stops playing and drops out of the system. The winner carries on playing and the average grade is now 1600.