I personally restarted to level 3 I think, I was at 7-8, because I had soooooo many reviews piled up after about a year long break. Plus the first time I learned them I didn’t really learn them as I cheated with the retry script 
I’ve been battling a pile of ~350 reviews for going on a week.
Every time I get it below 200 some life thing comes up and it jumps back up. It’s emotionally draining and sooooo frustrating to slog through 50 reviews, get a 30% accuracy, and it’s still not over.
Posts like these, and the support y’all hand out, really makes me feel like it’s possible and one day – maybe even tonight – I’ll get it down to 0 and I’ll keep it below low for a while.
1 Like
This is insanely high accuracy. Do you have previous Japanese experience, as in you’re reviewing things that you already knew?
2 Likes
I studied for 3 years in highschool, and then another 3 in University. I was like a B-/C+ student near the end though, mainly due to my lack of kanji and vocab knowledge
In the first ten WK levels I knew/remembered about 50% of the content, but at this point about 90% of it is “new”.
2 Likes
Ok that’s explains it. Also, maybe you are super human. I knew zero kanji before starting, and just recently managed to bring up vocab (my worst) to 90%. I was always all-A student back in a day, so having Bs was strange
2 Likes
My accuracy is like 80%, you aren’t alone lol.
wkstats also counts meaning and reading reviews separate and adds them together. For the total number of ‘reviews’ as it would have appeared in the box on the dashboard, you’d need to divide by 2 (roughly, not for radicals obviously).
So it looks like you’re pretty much on track
edit: just saw that @girakacheezer already mentioned this.
On that note: wouldn’t it just be easier to count the number of errors, multiply by 3 and add that to the 71008? Would give @Delamet and estimate of 119000 to end up on.
Probably not even that accurate, as Apprentice 1 items don’t drop and usually account for a significant portion of review errors.
All in all I’m not mad at @timfreilly 's original 100k+ estimate.
1 Like
On that note: wouldn’t it just be easier to count the number of errors, multiply by 3 and add that to the 71008? Would give @Delamet and estimate of 119000 to end up on.
Probably not even that accurate, as Apprentice 1 items don’t drop and usually account for a significant portion of review errors.
You’d want to take into account the chance of those reviews failing again, hence the geometric series/infinite sum. But you’re right in that it does assume that items would always drop 2 levels adding 3 reviews, when sometimes they’d only drop 1 or maybe even none if they’re at apprentice 1. Perhaps a factor of 2.5 would make more sense, to take that into account.
As for apprentice 1 items accounting for a significant portion of review errors, I think that would vary from person to person. Personally, my master reviews are the worst performing. I imagine that might change once I start seeing reviews of enlightened items though.
Ah, very true. I’ve only master reviewed a handful of level 1 items, so no experience with that. My apprentice 1 Kanji reviews are the worst at the moment.
This topic was automatically closed 365 days after the last reply. New replies are no longer allowed.