Here's an example of the power of cherry-picking to support an argument:
https://www.fflogs.com/reports/ptAZz...ing&source=403
You'll notice in this log that the whm and the sch had almost identical amounts healed, and that the whm cast med2 7 times and regen 6 times for a total of 13 casts. 2 cure 2s were cast along with 3 cure 1s for a total of 5 casts. If you were going to go by this data then clearly since medica 2 and regen were cast more times total than cure 1 and cure 2 then lily stacks should be tied to med2/regen, which may not necessarily be the case.