>>10907215
>Someone who kind of barely partially somewhat knows 10000 words because they've been reading is in a better position than someone who has 2000 words mature on anki
That's obviously true. Likewise, someone who kind of barely partially somewhat understands 10000+ words because he memorized 2000 kanji is in a much better position than someone who has 2000 words mature on anki. (And that's still under a faulty assumption that learning kanji in an organized manner is as slow as learning random words.)
>If you do plain kanji, the number of words you know is LESS than the number of anki cards you've gone through.
Now you're just being retarded.
Make a test. Take a random kanji, and search for it on jisho, common words only. Classify the results by how much of them you could understand from kanji alone.
Let's take 遅, for example. 10 hits.
Single-kanji words you can read right off the bat just with some basic intuition about grammar: 遅々, 遅れ, 遅い, 遅らす, 遅れる
Single-kanji words where your intuition will fail: none
Compounds you can read right right off the bat just by summing up their kanji: 遅咲き, 遅滞
Compounds you basically understand by summing up their kanji, but have a distinct meaning you'd probably need to check anyway just to be sure: 遅延, 遅配
Compounds you won't understand: 遅刻
Compounds where your intuition will fail: none
Total understandable words: 7 out of 10.
Looks almost too good, and admittedly there are many kanji that will give you more trouble, but it's not particularly special either. It's about normal.
And you can't even invoke the "but you can't pronounce/type them" card now, because you yourself praised "partially somewhat knowing". And even that can be countered just by pointing out how faster it is to learn all that once you know kanji.