The mysteries of Wanikani speed

Hey everyone! So glad to be joining the community. Seems like an awesome group of ppl here.

So, I just made it to level 6. I have about a 90-95% accuracy rate on reviews. I’ve studied Japanese for many years (finally committing to learning kanji properly and not forgetting it!!). This means that I know a lot of kanji at these initial levels. I know it’s going to get harder for me soon. My question is this: I’ve read a lot of forum posts about speeds that ppl go at, and it seems that for committed ppl, 7-9 days is pretty common. I level up about every 11-14 days. I do 15 lessons per day and never skip days. Why am I slower and what should I be doing to get faster?

To clarify: I don’t want to burn out (I’ve read lots about that happening). And I want to make sure I do everything ちゃんと. But I do want to take the JLPT N2 in the next year and I have a LONG way to go… so I’d like to get a bit faster. Maybe I need to aggressively use the lesson picker feature? Or maybe I should do Wanikani more? (I do it 2-3 times per day, spending about 1-2 hrs on it daily)

Thanks for everyone’s advice!

10 Likes

You level up once you’ve gurued 90% of the kanji for your level. You unlock more kanji by guruing the radicals for that level.

To level up faster, prioritize those radicals first (and it helps to start at least some of the kanji while you’re waiting on radicals). Then as soon as radicals are guru, do all the kanji lessons.

Fit vocab in as needed/when you have free time. It’s kind of troublesome to never do vocab because then you have a very large amount of lessons to do.

I think @NeoArcturus did ~25 lessons per day in the early levels and was able to keep a 7 day pace consistently.

4 Likes

Almost :sweat_smile: I was a bit lazy with my lessons until about level 13 and thus had to do 80 lessons within 24 hours shortly before leveling up lol, but according to my heatmap, I did about 26 to 27 lessons every day on average.

2 Likes

Yes, you would need to prioritize radical > kanji in the lesson picker to pick up the pace. You need to master enough kanji to level up, and to master enough kanji you first need to clear those radicals to unlock the rest of the level.

And if you really want to go max speed, you need to hit the reviews for those radicals and kanji as soon as they pop up again to master them faster. The first three times or so they’re only hours apart. The last time is two days apart. Even if you don’t try to hit those timers, that should set you to about 8 days or so to level up on average. (My pace)

3 Likes

thanks, this is helpful! out of curiosity, does it make a difference if you mess up kanji during the initial lesson quiz (right after you learn a lesson?). this doesn’t impact the SRS stages, right?

1 Like

No you can mess up the lesson as many times as you’d like. Only reviews matter.

2 Likes

thanks – i will definitely start prioritizing radicals. i’ve just been following what wanikani gives me every day but i guess that’s just got enough if you want to go fast.

2 Likes

Not a direct solution, but have you ever considered Anki? If you want to aim for N2 next year, grinding kanji in vocabulary context might be more beneficial :slight_smile: .

2 Likes

I assume you means 15 lessons per day, not reviews. You’ll never finish WK with just 15 reviews per day.

Each level has approximately 150 items, divide that by how many days you want to level up in and that’s roughly how many lessons you need to do level up

When I was on WK I was doing 10 lessons everyday and was leveling up every 2 weeks or so

3 Likes

Most higher level users forget that the early levels have more lessons. For example, OP just leveled up to level 6. Level 5 has 202 lessons. Level 6 has 199 lessons. Until you level up to 14, you need to do about 30 lessons per day if you want to level up every 7 days, assuming you are not skipping vocabulary. At 25 lessons per day, you will be close but will level up every 8 days instead.

2 Likes

Have you tried the WaniKani estimator? You can use it to estimate what your review load will be in the future based on how many lessons you do per day.

4 Likes

I’ve been leveling up every 8-9 days by always doing all my reviews as soon as I can get to them and doing about 25 lessons a day.
As others have mentioned, focusing radicals > kanji > vocab is important for speed. When I reach a new level I instantly do all of the radicals, then while I work on guruing them, I get the other new kanji in and once the radicals are on guru, I work on getting those kanji in as soon as works for me. Vocab I use to fill up my daily lessons once I run out of higher priority items. (After the radicals I often just use the new daily lessons and sometimes hand pick some kanji in, just to get them through before radical guru).

The most important part is doing all of your reviews though. If you want to avoid burnout, it’s better to hold off on new lessons but try to keep doing your reviews.
I’m currently at about 180 reviews per day on average, so with my lessons that makes about 200 items to go through every day and it takes me about an hour or two, I think. (I do it spread out and don’t track the time)
Currently I’m kind of doing the 0/0 Challenge, though I don’t participate in the thread.
So at a similar level and rate, you could expect a similar workload.

2 Likes

thank you @pembo i did mean lessons!

2 Likes

thanks everyone for this helpful advice. i didn’t know a wanikani estimator existed, so thanks @Marifly. and thanks @felixn for describing your process so far

@WeebPotato i appreciate the Anki suggestion. are you suggesting Anki since it has JLPT vocab sets, rather than the more random vocab wanikani provides? will wanikani get me there eventually…just slower? or will i find the vocab lacking?

3 Likes

If you can handle doing 15 lessons a day, you will be ready for N2 next year. Well let me say that if you are consistent with WaniKani you will at least be ready in terms of Kanji. Reading Speed and Listening Comprehension are much more of a concern at N2 level. And vocabulary size in terms of non-wanikani terms.

@Marifly Where have you BEEN? I’m almost at level 60 and it all started with your climb the sky tree thread.

2 Likes

I’m so glad that you’re almost there! I climbed too fast, fell down and took a too long break. Decided to reset and restart. Hope I’m a little wiser now!

4 Likes

I mostly suggested Anki, because it lets you pick flashcard sets with most common words and focus on vocabulary which you will need more than kanji on JLPT.

WaniKani would get you to a part of the vocab and all of the kanji, but first, it will definitely be slower than your pace and second, it won’t teach you all of the vocab you may need on the test.

1 Like

I think it’s well worth going as fast as you feel comfortable for the first 20 to 30 levels, then slowing down for the rest. You don’t have to maintain your pace all the way to level 60.

Personally I went “full speed” at one level/week until level 40, although in hindsight I probably should have slowed down earlier than that, around level 35 maybe, I was really burning out by the time I got to 40. And by the time you reach the 2nd half of the course you should really spend more time actually reading Japanese rather than doing WaniKani 3h/day.

The dirty secret is that in order to go fast you want to do things not ちゃんと at all. Very 汚い. Reaching level 30 quickly with 80% accuracy will generally be better than “only” getting to level 20 with 95% accuracy. For language learning quantity beats quality.

I cheat all the time with undo scripts, according to wkstats I have over 95% accuracy but it’s a complete lie, if I actually failed all the reviews that I miss I’d probably be around 80%. I don’t really care about that, I’ll have plenty of opportunities to revisit all that content when I encounter it in the wild. There’s no point reviewing a word again and again and again if you don’t even know if you’ll actually need it any time soon.

And that’s especially true for WaniKani vocab that’s pretty niche sometimes, and you have also no way to suspend vocab that you’re not interested in learning right now. I’d rather kick leeches to Enlightened and give them an other try in a few months.

4 Likes

According to the JLPT Chart at wkstats, in order to cover all of JLPT N2 kanji by using WaniKani alone, you would have to get up to level 51 by your deadline. [ETA: Important caveats below in comment by @pm215 : The mysteries of Wanikani speed - #20 by pm215]

So, first off, you need to consider if getting to level 51 by your deadline (using WK alone) is a) even possible at all with WK’s gated levelling system, b) idealistically possible, assuming you had unlimited time & energy, but maybe not perfect accuracy, c) realistically possible, assuming more realistic limits on your time & energy.

Even if a) and b) are possible, you need to also consider c). Given that you’re saying you want to get to N2 ‘in the next year’, and assuming that means ‘within 365 days’, then I’m guessing that technically a) and b) are possible, but I would strongly suggest that you consider how much stress and pressure you’d be putting on yourself when considering if c) is possible.

If you have a hard date for JLPT N2, and that is your primary goal, then I would suggest that – while WK may be a good, strong, baseline learning tool – you may want to consider supplementing it with a more customizable tool which you can use to specifically focus on drilling JLPT Kanji more directly and (perhaps) ‘efficiently’. At least ‘efficient’ in the sense of drilling for a specific JLPT test date.

There’s the tried-and-true Anki, which is very customizable, and certainly has plenty of JLPT-specific decks to choose from, so you could kind of ‘default’ to trying that out. It’s free, too, AFAI-can-recall. And the latest version has a new SRS system that is supposedly more efficient, too.

But there are also other more-WK-like systems, focused specifically on learning Japanese/Kanji, which I’m sure other folks here will know more about than I do. At least one of those will be customizable enough, but also user-friendly enough, that it would be competitive with both WK and Anki as a best-choice for cramming for JLPT N2 kanji.

As for if you need to work on JLPT grammar, then the best choice, IMHO, aside from direct study (like with a textbook or course or whatever), for SRS, is Bunpro. It is particularly good for grammar (hence the name, a play on words with 文法ぶんぽう), but it also has a new Vocab system, still in Beta, but getting better and better all the time, which also has Vocab Decks specifically for JLPT levels.

Depending on what you mean by “studied Japanese for many years”, then maybe getting to JLPT N2 would actually be very achievable. I would say that if you were starting from scratch (such as I was), then getting to N2 within a year would be super difficult and/or stressful/draining.

Heck, if you’re experienced enough already, then maybe even WK, with its gated levels, would even be enough. You could just focus on the Kanji and pick and choose which Vocabs you want to add in for the purposes of reinforcing those Kanji you find more tricky to remember. (You could use the Advanced Lesson Picker to accomplish that with ease, I’d imagine.) That might work!

Good luck!
:sunglasses: :+1:

1 Like

The big things to remember when looking at that chart are:

  • Directly tested kanji are a very small part of JLPT
  • There’s no “official” JLPT list of either kanji or vocab any more – the test might contain anything the test setters consider reasonable for somebody at N2 level to know
  • Knowing a kanji alone is not all that helpful – you need to know the meanings of the words it’s in
  • If you do encounter an unknown kanji or word in a reading question on the test there’s a fair chance you can make a good guess at the meaning anyway

For me, these things add up to it being a clearly better time tradeoff to study all the other things the test requires (grammar, vocab, reading proficiency) rather than pushing your WK level up that chart from a theoretical 60 or 70% coverage to even 80 or 90%, let alone 100%.

I also tend to feel that pure vocab study is in a similar “useful but don’t focus too hard on it” bucket – there are lots of words that might be on the test, and most of them won’t turn up in the specific exam you take. There are a lot fewer grammar patterns, so they’re much more likely to turn up on your exam. So grammar and reading and listening practice tend to have better expected payoff than kanji or vocab study

(But it’s also good to do some practice tests to see where your personal areas you need to work on are.)

3 Likes