Question regarding speed

So using the wkstats site I made some calculations about the relationship between the average number of new items per day and the time it would take to learn all 8804 items (not burn but at least guru) on the site, but I think my calculations are wrong based on people’s various posts about speed and what I average per level.

So my question is simple, how many new items per day on average, assuming 90-95% accuracy, would take to reach and guru the level 60 items in say, 1 year, 1.5 years, 2 years, 2.5 years?

For example, @jprspereira has written posts that say that he will reach level 60 in roughly ~365 days by using the site 1 hour a day, in three sittings (because of the way the SRS works in intervals of 12 hours, I think most of you are familiar with the post where he talks about it). I can’t figure out how that is possible. Wouldn’t just doing the lessons take more than 1 hour a day, let alone doing reviews? Or am I doing my math wrong and actually the number of new items per day you have to do is actually lower than I think (my time spent per level on wkstats doesn’t support that though, I’m even going somewhat slower than what my calculations would suggest).

And I’m really asking for people with experience, because its one thing to do the math, and a whole different to actually put it in practice.

Thank you.

1 Like

That depends on how you space out your lessons. You could spend an hour pounding through all of your lessons upon hitting a new level, but then you wouldn’t have to do lessons again until the next level. (I prefer spreading lessons out over a few days, btw.)

As for doing reviews in an hour a day… maybe. The faster you go the more reviews you have (though this can be modified by accuracy and typing/recognition speed) so for someone going max speed I’d expect to spend 1 hour as an absolute minimum.

Then again I don’t time my lessons, so I don’t really know how much time I spend in a day. But even averaging ~9 days a level, I feel like I’m spending more than an hour on reviews per day.

Are you assuming doing all lessons available in 1 row or something? I assumed 1 minute per item learned. Some items might take longer than that while others might be so straight forward that you just don’t spend any time with them (vocab that everyone knows like 俺、僕、猫、犬、私、勉強、etc).

One thing to take into consideration is that the amount of reviews you’ll get will increase the higher level you are. However, your speed also increases. Let me give you my personal example:

I remember thinking that a batch of 20 reviews was a lot of work at your level. I liked batches of 10 items or less. Now, I just completely refuse to do batches of 10 items because making the decision of doing them tires me more than actually doing them. The other day I checked my time and I did 280 reviews in 40 minutes (7 reviews per minute). This was unthinkable to me at level 5.

Of course, there are tons of ways to do your reviews. I could be doing them in a restaurant and being unable to focus because of the noise. That would increase the time I take to do them easily by 3x or more. The environment you give to yourself during the review session in also very important.

Also, you won’t get these crazy numbers until much later in your journey. Your level is not the only thing that increases, but your ability to be effective also does :slight_smile:

I think it’s not much about how others do it that you should worry about, but more about your specific situation. Is there anything in particular that is worrying you?

4 Likes

Thanks both of you for replying.

Exactly the opposite. Like I said, I’m talking about average new items per day, exactly in order to avoid doing all new lessons as soon as they become available.

So no concrete numbers?

1 Like

I don’t have enough skills to do some crazy simulations and turn them into graphics. But the simplest way to do it would be to divide total items by items learned per day.

8804 items/24 new items a day = 367 days. This includes all radicals, kanji and vocabulary. Of course, each takes different amounts of time and effort to learn.

8804 items/20 new items a day = 440 days.
8804 items/15 new items a day = 587 days.

I recommend separating the vocab from the radicals/kanji because they affect your progress differently. Radicals/kanji should be taken into consideration depending on the number of days you want to spend at each level. Vocab should be then spread throughout those days.

Another important question is why: why would one want to do WK the fastest way possible. After lvl 40, you’ll be learning less common kanji that you could very well learn through reading. At this point, your brain is a master in learning new kanji. The post you mentioned was meant to teach how to level up the quickest way, but that doesn’t mean it should be the only path to follow. A lot of people focus too much on WK and end up not spending enough time on learning vocab and grammar. If we talk about reading/speaking/writing, then everyone will tell you they don’t do enough of it. But oh well, not sure if this answers your question :v:

2 Likes

It has been my experience that my speed (of each individual session) has gotten alot faster as I have gotten used to the site. For me a 100 item review is about 15-20 min.

You can look at WKstats.com and it will show you what your average new item per day amount is. Have been doing 8 (and change) day levels and am doing 19.something new items a day.

so assuming 1 min per item for lessons (my exp is that this is faster as well) and two 100 review sessions that is under an hour a day. (average of course. unlock days are more, day threes less)

on the safe side say 9 day levels on average, that is a little under a year and a half to guru everything (but it would actually be shorter than this since the last ten levels are 4-5day levels as is level 1).

For the above to work out the whole way I have to stay dedicated and not burn out, but this is without ever scheduling specific times to do reviews or anything like that and maintaining 90-95 accuracy average.
Just when I wake up and before bed and no additional studying outside of lessons and reviews.

So all in all I am currently taking a pretty low stress approach. I think its a personal preference as to if finishing a month or 4 faster is worth the structure needed to pull it off.

I actually don’t want to learn WK the fastest way possible. I mentioned you specifically in my first post because you talked about doing a lot in very little time per day. Its the time per day (and energy) that I’m in extreme lack of, this is whats it about in the end, at least for me. If someone had unlimited time, they could go even faster than all of WK in one year (I have a friend that from absolute zero to taking N3 with a great score took him slightly less than an year, but he had pretty much all the time in the world to study Japanese [and he did it all self study with textbooks, no teachers or fancy digital resources], so he did). I want to find out how the little time and energy per day I have can be best utilized, min-max it in a way.

I did the same calculation as you, but putting it in practice didn’t give the expected results. I don’t know if its because its not corrected for 90-95% accuracy/time it takes to get an item from Apprentice to Guru/leeches/days where you don’t stick to the schedule 100%/etc, but it just didn’t add up the way one would think based only on that. Doing the same calculation, only not for all items in WK but for an average item number per level (at least for the first ~half of the levels, though from eyeballing it they don’t differ that much) brings a different, more accurate result.

But again, I might have just derped on the math along the way, or used wkstats incorrectly.

In the end the question still remains, what is the real average number of new items/time per day, with all possible obstacles along the way on the actual journey, after a person has dinged 60 in x, y or z time-frame?

1 Like

I think I understand what you are asking. The problem here is that you are asking 2 questions. The first is easy:
new items per day = 8804 / number of days it takes you to finish.

The second (time needed per day) is way harder. Do you have 80% or 98% accuracy? (that is going to have a big impact on your daily review load). Do you type fast? Are you reading outside of WK so the words you “know” come to you instantly or do you have to say the mnemonic to yourself every time you see a Kanji or think about items for 30 seconds before you can remember them? (thats gonna make a huge difference in the amount of time it takes to get through a review session of any given size), etc.

1 Like

This is exactly why I asked in the first place. Like I said, I did the math myself as well.

The answer to such questions of course is - sometimes yes, and sometimes no.

That’s why I was interested in the input from someone at the end of the journey (please no one take this as rudeness, all posts are welcome), because they can say what they did and we can see how much actual time it took them, corrected for everything life can throw at you in that time.

I think you’re overthinking. You can’t predict data you haven’t yet obtained. Getting a kanji that would allow you to level up wrong is totally different than getting a vocab wrong. The latter doesn’t influence the leveling up system while the former can be enough to delay you 3 extra days (aka getting it wrong at Apprentice 4).

Another thing to take into consideration is that WK is not the thing that will require more of your time. Reading, writing, speaking, learning grammar and vocab extra Wanikani will all take effort and time too.

I’d advise you to focus on building a habit out of Wanikani. Try to make all pieces of the puzzle fit together. You want WK to be as close as brushing your teeth. You don’t think about doing it. You get up and you just do it. This will save you mental energy and time to focus on the other stuff involved in learning Japanese.

1 Like

I don’t care for other aspects of learning Japanese at the moment. Like I said, I physically don’t have the time and energy for it.

That said, I picked up WK exactly because it would ease the rest in the future. I did a Japanese course some years back, and I found out the biggest hurdle for me was by far remembering the vocabulary (I actually didn’t even bother with kanjji for the vast amount of time time I was in the course). Not that grammar was a piece of cake, but there is a hierarchy that has to be in place for all things to progress smoothly. If one element, especially a foundational one, is not up to scratch, you are being slowed down by a lot. Its like missing days back in school. When you finally go back, its a lot harder.

It goes without saying that studying is somewhat like working out. You start by doing the maximum that would let you stick to your workout (which the heavy lifting part is usually 1 hour), with the best form you can, and most importantly, giving you the motivation to keep it up indefinitely. It doesn’t matter with what weights you start or are at the moment, as you progress you will naturally increase it without that being stressful or painful.

But figuring out stuff like sensible, proven average time/number per workout/sets/reps etc. is crucial. This is what my question tries to essentially find out, only for WK. And you need people with experience that have already done what you are doing to figure that out.

It sounds like what you really need is to just make your study habit the most efficient you can, and then just do as much as time allows and not worry about how long it will ultimately take.

To maximize your efficiency, using Ultimate Reorder helps, as long as you have the discipline to not abuse it (like letting your vocab fall behind). It takes 3.5 days to Guru your items. So, at the beginning of the level do ALL of your Radicals lessons first (using Reorder), and just spread out your Kanji over the next few days (e.g. divide the number of Kanji by 3, and do that many per day).

Then, at the halfway point of the level, the other (roughly) half of your kanji will unlock, and you’ll need to Guru them in 3.5 days. So, do all of your remaining Kanji lessons immediately, which leaves the Vocab. Spread out your Vocab lessons across the three days (e.g. divide by three, and do that many lessons).

That has the effect of keeping your daily workload relatively even.

Also, it helps A LOT to spend more time up front on lessons. Cycle through each group (e.g. of 5) repeatedly until you can recite them without hesitation. Knowing them better from the beginning means you’ll spend a lot less time thinking during reviews, so your reviews will go faster. Some people also use the Self-Study Quiz script on their newly-unlocked items to help cement them into your memory faster/better. It takes some time, but you can consider it as part of your lessons. It saves time later on the reviews.

There’s also a technique that a lot of people find useful: Any time you encounter a Kanji or Vocab, mentally (or verbally) recite the reading first, then the meaning, regardless of which question it’s asking. Without going too far into the details, doing reading-then-meaning can help you think in Japanese faster due to the way your brain works.

6 Likes

It is indeed way harder to memorize vocab when you’re not familiar with kanji. You can see this on Wanikani: if you know the kanji well, most vocab will come as second nature to you.

This varies greatly from user to user. Like I mentioned previously, I can speed through my reviews if I’m on my living room, without any background sounds, solely focused on doing them. Doing them in conditions where my attention will be divided (am I talking with someone? Am I listening to music? Am I comfortable? Did I sleep enough last night?) will dramatically change my ability to effectively execute.

So no. You won’t get the answer you’re looking for because there is none. You need to act on what you have in front of you right now, not on predicting whatever stat you wish to know.

You know what the max speed is. You then need to see if you can execute it by making a plan and then take action. That’s why I insisted on making a routine. The 3 times/day pattern that I wrote about focuses on doing the most in less time. You then will have to adjust to your own schedule. Is it leveling up every 7 days? Is it leveling up every 10 days? It depends on you.

Also, again, take into consideration that the first 30 levels teach you more common kanji. For example, at level 30 you’ll be able to recognize 86.48% of the kanji present on Wikipedia. By level 60, you’ll know 98.19%. This is a mere 11,71% bonus for the same amount of levels.

Since when have you been using Wanikani, if I may ask?

There seems to be some miscommunication going on.

Like I said, I have zero desire in doing WK the fastest way possible. I don’t care for 3.5 days for guring items, 1 week per level, 1 year to reach level 60 or whatever else.

I’m looking for the optimal way to use the tiny amount of time I have the best way possible. Note maximizing the effect of time spent per day, not shorten overall time.

And I’m not asking people to tell me what would work best for me. I know that people are different. I want to know what ultimately worked best for them.

I asked for the average number because it is an indicator with all factors included that had an effect while a person leveled, sometimes they were at maximum focus, sometimes they weren’t, sometimes they might have even missed days, what would the average of that look like then? Based on many such numbers of different people one can come up with a strategy that works for them.

Because doing the math in a bubble alone is not that helpful without real experience backing it.

As a side note, now in WK I like vocabulary the most, I find them easier than the kanji by themselves devoid of any context and actual use, I think I’m doing the real heavy lifting of actually learning the kanji mostly from the words. It would seem pretty counter productive to me to do only kanji and skip the vocabulary. Also as a philologist the studies on optimal number of new item retention are pretty conclusive. Even if people can power through ridiculous amounts of new words/phrases/concepts daily, their efficiency in getting them to the highest level of “learned” per overall time spent actually suffers dramatically that way. Its kinda like the myth of productive multitasking, people can look that up as well and get the overall idea, results aren’t that different. (Though people should keep in mind these studies are made in correlation with actual time spent on lessons and revisions for a given time-frame, like a week/moth/year etc, so obviously a person with more time can do a higher average number of “optimally” retained items than one with less.)

Edit: As a final addendum, to make things as clear as I can, my question was spawned by the difference that the number of average new items per day that I got from doing the math, and what people seem to be actually doing to reach the time I input as an example, don’t seem to match. Doing the math for lets say 1 year level 60 doesn’t give you the actual number that people do in order to actually achieve that (from my observation on the forum, which may be just incorrect assumptions). Correct me if I’m wrong. Do you guys on the higher levels think from experience that the math that was done already in the thread actually holds water in reality, did I just derp on the calculations?

This makes sense to me. Don’t see any fault in the calculation. As long as you reorder so that radicals are learnt as soon as they become available.

1 Like

I’d suggest a couple different ways. First is picking 1-2 times per day to do your reviews and consistently do all your reviews at those times (i.e. one review in the morning and another in the evening).

The other way would be to use the Ultimate Timeline script, find when a lot of your reviews are coming up for the current level and do all of your reviews then.
https://greasyfork.org/en/scripts/13475-wanikani-ultimate-timeline

The easiest way to do this is by making several learning sessions. I find it easier to learn 10 words in 2 sessions than learning 20 words in 1 single session. Of course, this might not always be needed. There are words that are more difficult than others. Sometimes I’ll breeze through 20 words and other times I’ll struggle with 9. The more you learn, the better you’ll know your limit and when to stop.

I think this happens because usually people just mention the number of words learned. Like I said, radicals and kanji are calculated differently because they influence how fast one levels up. Most people on the forums will probably tell you they do the radicals straight after leveling up and then spread the kanji through the first days (what I call as the 1st batch).

Just looking at my own raw numbers:

image

I did (on average) 18 lessons per day, for 488 days total (1yr 2mo).

At 1hr/day, that would only allow 3m20s per lesson, not leaving time for reviews.
I didn’t track my time-per-day, but I know I spent more than 1hr per day. My goal was maximum strength of memorization, so I did a lot of drills. Those drills made reviews go extremely fast, like 5-10 seconds per item (mostly typing time).

Looking at my total review count
(which is really low since I had high accuracy):
image

At 5-10sec per review, and 281 reviews per day, that’s 23-47 minutes per day on reviews. That number is probably relatively accurate.

2 Likes

Well, obviously I’m trying to work up a schedule/routine. That’s why I opened the thread in the first place. My life currently doesn’t permit doing that easily however, and finding why I get the discrepancy between my math and actual time spent on level will help me better utilize what little time I have, when I have it (which is rarely consistent), hence the opening question.

Like I said, min/maxing. And I need other people’s personal, subjective experience in order to do that. That’s why I’m interested in finding out on average what worked for them, so tell me your schedules (with concrete numbers)!

I do the lessons in the order WK gives them to me. I do all the radicals in one go when I reach them regardless of their number, since I find them extremely easy. And they come after I finish the last vocabulary item of the previous level. I also almost always have open lessons at any given time.

Thank you for your detailed post, it will be very helpful.

What is your average time spent on level on wkstats? How much is it different between early vs. late half?