I’m about a year in an on level 7 (which seems to last forever) clear my reviews every day & only do the lessons once a week.
I’m pretty consistently getting 83% correct (yes 83 - weird) & at least 1/3 to 1/2 the mistakes are fat finger typing - mainly “I” instead of “u” which is soooo annoying.
I’ve nothing to benchmark that with - is 83 a rubbish score should I be at or close to 100 every time ?
I personally think 83% is not a bad result at all. I usually had somewhere around 90%, but I think 80%+ is not bad either.
In any case, this is not a test or exam – here when you fail an item, it’s not an indication that you didn’t study well enough. Instead, it’s an indication that you should study that item more, which is what SRS is for. So, my recommendation would be to just relax and do reviews at your own pace and don’t bother too much about the efficiency. Then again, different things work differently for different people…
In any case, best of luck with your studies! 
Hello!
Look, I am not that well versed, but I do get around the same percent as yours, and I think it’s okay! 83 is nothing to scoff at! Plus, I personally find doing a lot of mistakes to be part of the process, and actually help you in ‘drilling’ infromation in! Which really isn’t your case, because, again, 83 is nothing to scoff at
If it helps here is a chart where I track my lessons
Though, I wouldn’t personally worry so much about the percentage as much as the amount of lessons if you feel bogged down
I personally do reviews everyday, and I find wanikani to be built for such a flow; Might I ask, why you choose to do reviews once a week? I find that to be rather slow, and I think you also share the same sentiment, if only subconsciously
Hope it helps in any way!
I think school really does a number on us, we all seem to need some qualitative boundary between success and failure, like 69% is bad or something.
As long as you’re learning and remembering more kanji than you knew last week or last month or last year, you’re improving! And improving is the only important measure.
Well, yes and no. I think if your accuracy rate is significantly lower than most people’s (not that that seems to be the case here) then that suggests that it’s worth looking at why. The answer to “why” might turn out to be something you can’t change or don’t want to change, but it’s also possible that it’s a pointer to something you could improve in how you’re learning and studying.
In particular if the problem is “silly typos are lowering my hit rate” then there are well established answers in the form of using a double-check script or one of the unofficial mobile apps to get “undo” functionality.
Why do you do lessons only once per week? That sounds like you are deliberately slowing yourself down. How many lessons do you do each time? I think it would be better to do 1 lesson every day than 7 lessons once a week.
Regarding percentages, in my experience around 90-95% helps reduce review pile significantly over time. Around 80-90% is good enough to keep the reviews in check. If you see yourself frequently going below 80%, your reviews will pile up. It’s more about how much more effort you need to put into it to stay on top of reviews. But it makes sense - lower accuracy means you need to study it more.
I agree that you shouldn’t worry unduly about your percentage, but for what it’s worth, you sound about on par with where I was at that level. I think I had about 90-95%, but that was because I used undo functions when I had a legitimate typo. Otherwise, I’d have been in the 80s, like you.
To give you a heads-up, my percentage has dropped since then, so don’t be surprised if you notice the same trend. I think it’s just a matter of saturation beyond a certain level; the first thousand lessons mostly seemed distinct and simple, but the more I see, the harder it is to remember them all, which shouldn’t be surprising.
Also, I try to do 90 lessons per week. Though I don’t always keep up that pace, I notice that if I let my pace drop too far, my recall actually suffers. I assume that’s because seeing new vocab words will reinforce the constituent kanji, so moving through the vocab list more quickly actually helps me solidify the earlier lessons. It’s not a race, and everyone should go at a pace that works best for them, but I thought I’d offer that perspective, for what it’s worth. I think it’s impressive that you’re keeping your accuracy up at a slower pace.
I wonder what the experience is like for the max-speeders.
You could probably achieve similar benefits (plus additional advantages) by doing a lot of real-world reading, even if progressing more slowly through WK, but I don’t have enough grammar yet to read that much.
Do you review on a touch-screen device, or a desktop/laptop computer with a full-sized keyboard?
Many people find their accuracy is significantly higher on desktop than on a smartphone, enough to make it worth some loss of convenience. Or (as mentioned above) you can consider one of the third-party apps or undo scripts, if you feel you will not misuse it.
To burn an item, you need to hit streaks of at least three correct reviews in a row. (Otherwise, the item doesn’t make forward progress through the SRS levels, since a single wrong answer knocks it back at least two steps, at the later stages.) At 80% accuracy, the chance of any three given reviews being correct is only 51%. At 90% accuracy, it’s about 73%. You’ll see much more consistent burns if you can get closer to 90% accuracy.
(The above math is using a simplified model where every review has the same chance of success. Reality is more complicated, but the overall effect is similar. Also, if your accuracy stats are counting “reading” and “meaning” reviews separately, then you actually need a streak of six in a row to advance an item.)
