I don’t get it. You don’t go particularly fast or do a lot of lessons at once so what’s making up these numbers? But more importantly, are your review sessions more manageable now that you don’t have so many leeches hanging over your head?
And wait. You have nine Anki decks going?!
Which remind me, on the N2 two weeks ago 参考 appeared (more exactly 参考書) and I could confidently answer because I saw it on your study log that you updated just the day before. Thanks! 
Guru leeches.
Out of my (currently) 67 Apprentice leeches, only a portion of those are included in my “auto-pass” script. (Maybe I should add some more.) But I’m still doing all Guru reviews, and that’s where I have another 271 leeches. That’s aside from my Guru and beyond that are no leeches, but I end up forgetting from time to time.
Sometimes yes, sometimes now. Today I’ve had 102 reviews totals, but I think I had six auto-pass, meaning I’ve manually done 96 reviews. I’ll have 11 more come up soon to round out the day.
The main issue is that because I strain myself trying to remember these cards on reviews, only to fail, it’s time consuming and mentally exhausting. After having a 60-card review session, it’s harder to get to (for example) my daily manga reading.
There are two types of deck:
1) Imported decks of common words.
This is just review to see which common words/kanji I recognize by sight (no furigana), and these cards get suspended (burned) after getting them right a few times. In all, I think I started with about 2,500 cards cards, and I’m down to about 1,300 left (most as “lessons”). There’s also overlap of some of the same words between both decks. Since I know well most of the words I’ve started reviewing in these decks, they go very fast.
2) Decks I’m building as I go. These ones start empty, and I build them over time. For example, the Names deck only has 10 cards in it. My mined sentences decks have only a total of 202 cards (at least two of which auto-suspended as leeches).
Overall, I’m spending about half an hour a day doing Anki lesson+reviews, and that’s typically about 200 cards per day. That’s not counting time spent creating sentence cards for common kanji I’m re-learning/improving my recognition of.
I keep getting this one wrong in my Anki reviews =(
Aside from having trouble remembering how 参 is read (my focus for this card is meaning, so I’m allowed to show the furigana without failing it), it’s a word that I don’t encounter often. Thus I never remember the meaning. I do have a sentence to go with it, but even then I forget.
But, I think I’m starting to have memorized the shape of the example sentence! So, soon I should start to remember the meaning of the vocabulary word just by seeing the general shape of the sentence, before I even see the word. (That’s a very bad thing, though. It’s my weak point with sentences on cards.)
I think it’s safe to say that since you’re working on customized decks outside of WK, it might be a better use of your time to auto pass some more. Can you tell that I’m all for lightning the load when mental exhaustion sets in? Because I am. I’ve been dealing with burnout myself and have had to decide what making progress looks like while using as little energy as possible.
I was thinking about this earlier today. Which path do I want to take?
I know it’s silly and meaningless, but I kind of want to avoid breaking my review streak:
I think as I get things a bit settled in Anki, I should look at which kanji/vocabulary are WaniKani leeches and show up enough in immersion material that I should give them attention in Anki. Then see if that helps slowly lighten the load in WaniKani.
That is quite the streak! I generally only use vacation mode for the rest of the day after doing my morning reviews on Thanksgiving. Totally random, I know. And while I can’t say that I haven’t felt tempted lately, I’ve been able to push it off for another day and another…
I’m wondering in your case, since the desire to work on your Anki cards seems to be where your heart is right now, if it’s worth revisiting doing a strategic reset to lower your review numbers. It looks like auto- passing has helped somewhat, but as you wrote above, the drop in number hasn’t been significant enough to deal with mental fatigue and
which is pretty important to you.
Anyway, in terms of shifting priorities, where you’re leaning more towards prioritizing Anki over WK, as the CFitz who is always trying out new systems in language learning, I say go for it. Sometimes we need a refresh to make progress! And you’re not going to lose anything by focusing on a new tool over the one that you’ve been using for a long time because maybe that one isn’t serving you as well as it used to.
I feel this 
Ah, yes, that should be on the list as well:
I know some people are against doing resets, but I’m very supporting of a reset in certain circumstances. I’d say anyone with a lifetime account (me) who’s struggling with leeches (me) to the point that they’re not able to take on any new lessons (me) who is too stubborn to take a break from daily reviews (me) is a perfect candidate for a level reset.
It’s absolutely an option on the table. But I do want to try a few things first to see if I can find an alternate solution. Failing that, I’ll highly consider a reset.
This chart is just my leeches, not my overall cards. By dropping down to 20 (for example), I’d be removing over half of my leeches. This chart doesn’t tell which stage each one is, so this doesn’t tell if they’re leeches that I’ve been getting right (closer to Burn) or wrong (often in Apprentice).
I don’t know whether a reset erases all progress on cards (being demoted to Apprentice 1), but that could still be worth it if it meant spending time elsewhere in the short run to get the leeches under control. I would just need to ensure that when I return to doing WaniKani lessons, I don’t have the leeches building right back up again.
I did start yesterday with:
My highest frequency kanji that’s a leech is…待. The worst part is that I learned very well the difference between 待 and 持 probably around 2015 or so, and yet here it is among my leeches =(
I decided to start with 代 instead. I mostly get the meaning wrong in reviews, so I need to keep in mind how the meaning “substitute” applies for words I added to Anki:
Let’s see if I can make progress learning Japanese, one kanji at a time =D
Then there’s this one:
This is one thing that really helps with the Migaku kanji add-on. There, it adds common vocabulary words, as well as words from my other cards, to show when reviewing a kanji card:
It should be trivial to implement the same into WaniKani reviews, for someone familiar with with creating WaniKani userscripts. For me, I’ve settled for writing a script that looks at my pending reviews, and creates a web page for the pending kanji:
It’s something I can have up when doing kanji reviews. I haven’t actually started using it yet, but I plan to today. (And I may still need to tweak it to hide words I haven’t started learning yet. At the moment, I only filter out higher level words.)
I figure the worst that can happen is I become reliant on seeing a kanji in a word to know its meaning/reading. But Japanese isn’t kanji, Japanese is words (which just happen to be written using kanji). Thus, recognizing a kanji only when it’s in a word might be okay.
One potential downside is that it may make it more difficult to guess the meaning of unknown words using known kanji (or even to recognize that they’re known kanji). Well, I’ll give it a try and see how things go!
Thus far, I’m getting amazing results with this method.
I realize that 63% might seem like an amazingly terrible score to most people, but my usual score on leech kanji reviews is closer to 33% accuracy. By having vocabulary available, I’ve nearly doubled my accuracy.
The main issue was the ones where all of the kanji’s vocabulary were burned and forgotten, so I didn’t recognize any of the vocabulary words. I may need to unburn some vocabulary, or add words for the kanji to Anki.
if you really found a way to make the leeches die!!! you’ll be a hero!! a script you will need to release hehe
The main obstacle right now is “What is this vocabulary? I’ve never seen it before in my life! Oh, WaniKani says I burned it last year.”
For a lazy person like me, the things standing my way for creating a userscript:
That said, the WaniKani API does allow querying a kanji for the vocabulary words it’s used in, and from there querying the vocabulary words for their levels (and probably learning/burn status). For someone experienced in writing WaniKani userscripts, I would think it would be easy enough to write (maybe?)
still bet you could cobble something simple together 
really will depend if it works long term right?
If something were cobbled together, let’s say in an hour or two (of mostly learning how to do things), it might look something like this early on:
And from there, I’d probably need to wait until I had enough reviews pending that I could implement transitioning from one review to the next, as well as handle kanji versus non-kanji reviews showing up.
But one probably shouldn’t cobble things like this together until after giving enough time to determine whether this method even helps any for kanji recognition during reviews in the long-term.
cobbling is good!!! looks nice… 
Somehow I managed to get 17 out of 18 leech kanji correct in this afternoon’s review session. Having the vocabulary show definitely helps me recall the reading and meaning of the kanji.
Reviews take quite a bit longer, as I’m looking over the vocabulary words, trying to remember their reading and meaning, and then trying to remember which reading WK wants (kun vs on), and then trying to remember what keyword the kanji has. But it’ll be a time saver in the long-run versus constantly getting them wrong in reviews. (Although auto-passing them previously was nice and fast!)
About 30% of my overall leeches are kanji. It’d be great to see my leech count slowly go down by even that much. But that may be premature wishful thinking.
If I have another good leech kanji review session or two, and I may just clean up the code for release. (But then that opens things up for feature requests that I’d be too lazy to implement. “How about adding a configuration screen with an option to hide vocabulary I haven’t done the lesson for yet?” “What about only hiding vocabulary from higher levels?”)
Whether this helps or hinders (or neither) long-term kanji recognition…shall remain to be seen.
you could always enlist the help of some other super scripters on here…maybe someone else would be willing to maintain it … (cuz that’s probably the worst part)…of course when it’s breaks its all WK’s fault anyway for changing stuff w/o ask you the script writers first 
My favorite dialogue at work:
Coworker: “Chris, this program you wrote isn’t working. Can you take a look?”
Me: “Hmm, it appears the web team made some unannounced changes to the web site we pull data from. My web scraper can’t see the content anymore.”
(Except there I get paid, so I’m not lazy about fixing things.)
Well, I’ve done gone and posted a userscript.
So far, so good with using my userscript for showing vocabulary underneath kanji during reviews.
For about half of the kanji in my review just now, I stared at only the kanji for five to ten seconds, contemplated on whether I’d ever seen that kanji before in my life, and then tried to figure what I may have mistaken it for in the past to help with knowing what isn’t the right answer.
When I finally looked at the vocabulary words, I (usually instantly) knew the correct reading/meaning.
I now have a couple of milestones to aim for:
1. Get Apprentice kanji to 0.
2. Get Guru kanji to 0.
should probably have installed this before doing the last 3 reviews which failed them all…oops…next hour will find out how well it helps make leeches die… super looking forward to this!!!
