I believe I have figured out the BMI math for the new fitness age. First off, the BMI it wants you to reach is the exact midpoint of the "healthy" BMI range. This is a little silly as the entire reason BMI ranges exist is that your weight divided by your height is a good rule of thumb but in no way perfect for what your ideal weight should be. Granted, for me personally, said midpoint is actually within a pound of my ideal weight (based on advanced medical scans for accurate body fat percentage and an athletic amount of body fat). Still, that's not going to be the case for many people so perhaps letting people set their own ideal BMI at least within the "healthy" range would probably be an improvement.
Anyway, the other bit of BMI math appears based on the study mentioned in the article: How BMI and Weight Gain Shortens Your Life Expectancy (menshealth.com). It suggests that 1 BMI unit overweight reduces your life expectancy by 7 months and Garmin appears to be moving your fitness age by 0.5 years for each BMI unit. The study link is broken and I haven't had time to search for it via the university database, so it is unclear if there are any other things to consider in the regression equation though it claimed to remove all genetic influences based on the large sample of genetics and parental life expectancy. But from what I've seen, this seems reasonable as far as the gamification of fitness goes.