I nearly dropped this at chapter two. I’m glad I didn’t, but “it gets better” is the least useful thing one reader can say to another, so this is the version with the actual numbers attached.
The first checkpoint lands at 42/100, and that score is honest rather than a hedge. What the opening asks of you is patience with a story that refuses to explain itself — the art is deliberately disorienting, the pacing withholds almost everything, and there is no early signal telling you that any of it is on purpose. Two chapters in, I genuinely couldn’t tell whether I was reading something carefully built or something confused.
That distinction took me until roughly chapter fourteen to settle, and it is the whole reason I score as I read instead of once at the end.
Why the number moved
Forty-two at chapter two, seventy-nine at chapter forty-five. Thirty-seven points is the largest move on this site, and the movement is not spread evenly — most of it arrives between chapters fourteen and thirty, when the thing the early chapters were witholding finally starts paying out.
Two things climbed together and one lagged. Payoff went from 6 to 16, which is the clearest signal that the setup was real. Characters went from 10 to 17, which surprised me most: a story this invested in atmosphere usually treats people as scenery, and this one doesn’t. Pacing only moved 7 to 14 and stays the weakest axis. It is still slow. I have simply stopped reading the slowness as a fault, which is a different thing from it being fixed.
What it does well
Restraint, mostly. There is an obvious version of this story that reaches for shock, and this isn’t it — the dread is built from what you’re allowed to see rather than what you’re shown. The art is doing real structural work, not decoration, which is why the art score has led the story score at every single checkpoint.
And it is one of the very few things I’ve read this year that is genuinely unnerving without ever being cruel about it.
Where it could still go wrong
The pacing score is the one carrying the risk. Fourteen out of twenty at chapter forty-five is fine, but if the next arc spends another fifteen chapters on atmosphere without forwarding anything, this becomes a series I admire rather than one I recommend. The trajectory is upward right now; the reason I keep checkpointing is so I can see the exact chapter where that stops being true, if it does.