The Standings → Claude Sonnet 5.5 (medium)
Anthropic
medium reasoning effort · via Claude CLI · judged by Claude Sonnet 5 · not the setting shown on the standings; also run at xhigh78.3 high74.1
Results from run ‘first’: ten items, ten verdicts.
Claude Sonnet 5.5 (medium) scores 72 on craft, 9 points above the field’s 63 and 21 points short of the authors themselves. Its shape differs from the field’s: relatively strong on restraint, relatively weak on the author’s voice and sentence-level prose. On the mechanical checks it kept to the brief’s word range on 5 of 10 items, reproduced a 25-word run of the real book verbatim and leans on stock phrases more than most.
The record: 10 items · 10 verdicts · 0 errors · 1,820 words on average · est. generation cost $0.32
Eight dimensions, each 0–100, averaged over ten items; the author’s real chapter was in the room every time.
Craft by dimension, 0–100 · judged by Sonnet 5 beside the author’s real chapter
The authors themselves average 93 on the five writing dimensions, the field 63, this model 72. Hover or focus a bar for the field and the authors on that dimension.
The judge’s one-sentence best quality and worst flaw for each chapter, written for a general reader.
Best The final solitary walk home at dawn is beautifully observed, with quiet, concrete details of Paris waking up that really capture the book's mood.
Worst At one point Jake spells out his feelings for Brett in a longer internal passage ('I thought that I would like to sit there and never say anything...') that explains more than the spare, show-don't-tell style used elsewhere, which softens the effect of the plain sentence that comes later.
Best Elizabeth's throwaway line about keeping the carriage while giving up the hot-house is a small, quiet joke that shows her real priorities without the narrator ever spelling them out.
Worst Mr Shepherd's own retrenchment plan (timber and mortgages) is a bit generic and doesn't give him as distinct a voice as Sir Walter and Elizabeth get.
Best Robert's teasing banter with Adèle on the walk back captures his usual light, self-regarding charm very convincingly.
Worst The chapter never actually gives the lady in black her required first words to Edna — she's mentioned only as having already left the beach.
Best The Yorkshire servant's dialect and Mary's blunt 'I don't understand you' feel natural and true to how a foreign-raised child would actually react.
Worst Mrs. Medlock locking Mary's door and the pale eyes that glow and vanish in the dark push the chapter toward a spookier, more sensational tone than the quiet unease of the original book.
Best The wife gets real dialogue and a clear point of view of her own, so her worry feels like a person's rather than a plot device.
Worst The wife calls the narrator 'Nils,' which breaks the book's whole point of keeping him unnamed and reads like an accidental slip rather than a real choice.
Best Lord Emsworth's contradiction about whether he actually asked Eve about her qualifications ('Did I? So I did.') is a lovely, understated bit of his trademark absent-mindedness.
Worst The scene where Walderwick discovers the umbrella is missing is muddled — he seems to think he still has his own umbrella at home even while holding the note about the borrowed one, which blurs the joke instead of sharpening it.
Best It hits every required plot point cleanly and without confusion, and the closing image of the open back door with no Rosanna coming through it is a nicely understated close.
Worst The explanatory letter is signed by an invented, unexplained figure ('Samuel Blake-Gray, Clerk') who shares Franklin's surname and whose employer is never identified, so the device that justifies the whole chapter feels bolted on rather than grown from the story.
Best The final telegram is a genuinely unsettling reversal, turning a page of reassurance into a reason to be frightened, which is exactly the kind of hook that makes you want the next chapter.
Worst The dinner is accidentally placed on the wrong night, and the closing telegram's revelation arrives almost too fast to be believable given how slow post to Transylvania is shown to be elsewhere.
Best The unnamed messenger with the cauliflower ear delivering Thaler's invitation is a good small, concrete detail that makes the gang's reach feel real without overplaying it.
Worst The ending turns into a flat list of the detective's deductions ('If the check had bought something, I didn't know what. If it had been a payment...') that reads like a recap instead of a scene.
Best The dry-washing detail that catches Inglethorp's story in a lie is a small, plausible, well-observed piece of detective work that feels exactly like something this book would do.
Worst The investigation moves through its clues so briskly and tidily that the big moments (Inglethorp's return, the object Poirot finds) don't get room to feel tense or strange, and the very last beat fumbles its own ending by having Poirot pocket the object instead of leaving it out where the reader can sit with the mystery.
Headline, compliance and craft for each chapter, 0–100.
| Item | Headline | Compliance | Craft | Words |
|---|---|---|---|---|
| The Sun Also Rises item 021 · after ch. 3 · Ernest Hemingway | 79.9 | 100 | 80 | 1,688 |
| Persuasion item 001 · after ch. 1 · Jane Austen | 78.5 | 100 | 78 | 2,491 |
| The Awakening item 016 · after ch. 7 · Kate Chopin | 74.8 | 95 | 79 | 1,394 |
| The Secret Garden item 013 · after ch. 2 · Frances Hodgson Burnett | 73.5 | 100 | 74 | 1,776 |
| The War of the Worlds item 006 · after ch. 1 · H. G. Wells | 69.5 | 95 | 73 | 1,744 |
| Leave It to Psmith item 022 · after ch. 3 · P. G. Wodehouse | 67.2 | 95 | 71 | 2,099 |
| The Moonstone item 011 · after ch. 8 · Wilkie Collins | 67.0 | 95 | 70 | 2,042 |
| Dracula item 015 · after ch. 5 · Bram Stoker | 65.1 | 85 | 77 | 1,338 |
| Red Harvest item 023 · after ch. 2 · Dashiell Hammett | 58.8 | 100 | 59 | 1,735 |
| The Mysterious Affair at Styles item 017 · after ch. 3 · Agatha Christie | 54.7 | 95 | 58 | 1,894 |
Words written per item, against each brief’s maximum.
Word count per item vs. brief maximum · run ‘first’
Five of ten items landed inside their brief’s word range.
Each verdict names the chapter’s single worst flaw, with evidence. Quoted verbatim from run ‘first’.
The investigation moves through its clues so briskly and tidily that the big moments (Inglethorp's return, the object Poirot finds) don't get room to feel tense or strange, and the very last beat fumbles its own ending by having Poirot pocket the object instead of leaving it out where the reader can sit with the mystery.
At one point Jake spells out his feelings for Brett in a longer internal passage ('I thought that I would like to sit there and never say anything...') that explains more than the spare, show-don't-tell style used elsewhere, which softens the effect of the plain sentence that comes later.
The explanatory letter is signed by an invented, unexplained figure ('Samuel Blake-Gray, Clerk') who shares Franklin's surname and whose employer is never identified, so the device that justifies the whole chapter feels bolted on rather than grown from the story.
The scene where Walderwick discovers the umbrella is missing is muddled — he seems to think he still has his own umbrella at home even while holding the note about the borrowed one, which blurs the joke instead of sharpening it.
Automated screening results across the ten outputs.