I have a suggestion about how spaced repetition is working (with the possibility that some of what I’m seeing may be due to the quirky word behavior I’ve commented about in other threads).
Right now, when a character that is used in many words is introduced (ex: guo2) that character is then used in many other words (ex: ying1guo2, ying1guo2ren2, zhong1guo2, etc etc). It does not appear that those word usages are factored into when the character guo2 is repeated. So, for characters used in words, you can get a lot of repetition such that you never need to test the character itself. Marking it as too easy could help with that but it would be nice if the repetition algorithms accounted for it.
I would prefer new characters, when introduced to cycle through multiple times in short order maybe interspersed with more recent learned characters. I’ve noticed from the repetition of characters through words that I can learn the characters much faster that way, but many new characters are not also included in words so they don’t get as much immediate repetition (again, this might be the Android app misbehaving - so hard to tell). It’d be great if those new characters got a lot of repetition right away to get them learned quickly. Maybe it is supposed to work that way, but it doesn’t seem to.