Well, that’s a wrap.
Back on June 11th, I started an “epilogue” to my previous Wanikani challenge. With my review count now back up to 1569, I wanted to see how many I could burn. Now, 54 days later, I’ve finally hit 0 for the first time, thus ending the challenge.
Over those 54 days, I did a total of 3724 reviews and spent a total of 15 hours and 9 minutes doing reviews. Of those reviews, I missed a total of 1135 reviews, for an overall accuracy rate of 69.52%. However, that obscures the underlying patterns.
My main goal was to see how many items I could burn. At the start of the challenge, I had 901 items at the Enlightened level, and thus 901 items I could possibly burn. Of those, I only actually managed to burn 608, meaning I had a 67.48% accuracy rate on burn reviews.
Here’s a graph showing the number of items at each SRS level over time:
As you can see, the number of burns went up quickly at first, but tapered down over time. This is because at first, almost all the items up for review were Enlightened items that could be burned. But over time, as I either burned or failed all of them, all the possible burn reviews went away and the leeches made a higher and higher proportion of reviews.
Although I only finally hit 0 reviews this morning, my last successful burn was nearly two weeks ago on July 22nd, and the last (failed) burn review at all was on July 26th.
By the end, I was only reviewing items that I’d previously missed before, often many times (since when you miss them, they stay at Apprentice and keep coming back up).
You can also see this pattern in my speed and accuracy graphs. Here’s a graph of a five-day weighted rolling average of my review speed:
I started out very slow due to not remembering anything after being away from WK for over four months. Plus most of the items I was reviewing were high level burn reviews. Over time, seeing the reviews helped jog my memory, and also over time, more and more of the reviews were leeches that I’d already reviewed and missed recently, making things much easier.
This also lead to a corresponding increase in accuracy over time as well: (also five-day weighted average)
Anyway, I guess that’s it. I originally thought about doing a second epilogue four months from now to try to burn more items, but I decided there isn’t much point because over time, more and more of the items will just be impossible-to-remember leeches.
Therefore I do not plan to return to Wanikani unless they add new lessons again, and it’s been a long time since they did that, so I doubt it will happen again. Therefore, this seems like it is really the end. So long and thanks for all the 魚.