The Standings → Claude Opus 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 xhigh77.4 high74.9
Results from run ‘first’: ten items, ten verdicts.
Claude Opus 5.5 (medium) scores 76 on craft, 13 points above the field’s 63 and 17 points short of the authors themselves. It sits close to the field on every dimension; its own best is the author’s voice at 78, its own worst restraint at 75. On the mechanical checks it kept to the brief’s word range on 6 of 10 items and reproduced a 72-word run of the real book verbatim.
The record: 10 items · 10 verdicts · 0 errors · 2,536 words on average · est. generation cost $1.22
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 76. 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 It picks up the scene exactly where the source chapter left off, with Brett's line about being miserable answered right away, which makes the opening feel seamless.
Worst A few moments, like the violets, spell out the emotion a bit more plainly than the source material ever would, softening the story's usual reticence.
Best Anne's big speech and the family's stunned reaction to it feel earned and true to her quiet, put-upon character rather than staged for effect.
Worst The narrator explains exactly why Elizabeth backs Anne (fear of meeting Mr Elliot in town), which spells out a motive that would have landed better left implicit.
Best The small human and comic touches (the doctor gossiping on the doorstep in front of the whole street, Poirot's moustache 'elaborately curled as though he were bound for a garden party') make the world feel textured and alive.
Worst It's competent and clean but a bit safe — the big confrontation with Alfred and the room search go by smoothly without ever really raising the emotional temperature.
Best The detail of the mother pony and her foal blocking the road is a lovely, tender, concrete touch that makes the moor stop feel specific and alive.
Worst Large chunks of the house-arrival scene (the oak door, the dim hall, 'a very small, odd little black figure') are copied almost word-for-word from the real published chapter rather than newly written, so that section isn't really this writer's own prose.
Best It gives real psychological depth to Ogilvy, showing him talking himself out of a doubt he's clearly having rather than just being told he's wrong, which is a sharp, very Wellsian piece of characterization.
Worst The falling star is timed as happening near the end of the ten nights of Mars-flashes, but the source chapter implies the object should still be weeks away from Earth at that point, so the timing doesn't add up; the crowd's reactions during the burn scene ('A woman cried out') are also fairly stock rather than individualized.
Best Small physical details, like the odd lightness of the sealed packet and the realization of how much a key can weigh, make the scene feel lived-in and real.
Worst Lady Verinder's big speech forbidding her brother's name sounds like a formal proclamation rather than something a real person would say at teatime.
Best The bank scene and the widow's guarded, evasive answers sound like believable, tight hardboiled dialogue that keeps the mystery closed rather than explaining it away.
Worst The chapter ends on a flat, repetitive note, rehashing the same thought about the missing check instead of pushing the story forward.
Best The scene where Psmith and Eve share the umbrella and talk on the way to Shaftesbury Avenue is warm, funny, and reads like real Wodehouse banter.
Worst It stops to explain Freddie's Oxford and chorus-girl history in a lump instead of just showing Lord Emsworth's panic, which slows the funniest moment down.
Best The closing telegram is a genuinely gripping gut-punch that instantly recasts everything the reader has just been reassured of.
Worst The Renfield mystery and the forged-letter mystery both get tidy, spelled-out explanations a little too quickly, which lets some of the air out of the suspense before the final twist arrives.
Best The lady in black's short exchange with Edna is quietly strange and watchful, true to a character who has barely spoken before.
Worst Robert's remark to Adèle reads like polished literary criticism rather than something a playful young clerk would actually say out loud.
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 | 80.4 | 100 | 80 | 2,220 |
| Persuasion item 001 · after ch. 1 · Jane Austen | 79.6 | 100 | 80 | 2,579 |
| The Mysterious Affair at Styles item 017 · after ch. 3 · Agatha Christie | 76.0 | 100 | 76 | 2,542 |
| The Secret Garden item 013 · after ch. 2 · Frances Hodgson Burnett | 75.0 | 100 | 75 | 2,565 |
| The War of the Worlds item 006 · after ch. 1 · H. G. Wells | 74.1 | 95 | 78 | 2,629 |
| The Moonstone item 011 · after ch. 8 · Wilkie Collins | 73.8 | 100 | 74 | 2,873 |
| Red Harvest item 023 · after ch. 2 · Dashiell Hammett | 73.8 | 100 | 74 | 2,440 |
| Leave It to Psmith item 022 · after ch. 3 · P. G. Wodehouse | 73.5 | 100 | 74 | 3,390 |
| Dracula item 015 · after ch. 5 · Bram Stoker | 73.3 | 95 | 77 | 2,456 |
| The Awakening item 016 · after ch. 7 · Kate Chopin | 72.6 | 95 | 76 | 1,663 |
Words written per item, against each brief’s maximum.
Word count per item vs. brief maximum · run ‘first’
Six of ten items landed inside their brief’s word range. The Secret Garden and The War of the Worlds and The Moonstone and Leave It to Psmith ran over by 165 and 29 and 73 and 590 words respectively.
Each verdict names the chapter’s single worst flaw, with evidence. Quoted verbatim from run ‘first’.
The falling star is timed as happening near the end of the ten nights of Mars-flashes, but the source chapter implies the object should still be weeks away from Earth at that point, so the timing doesn't add up; the crowd's reactions during the burn scene ('A woman cried out') are also fairly stock rather than individualized.
Large chunks of the house-arrival scene (the oak door, the dim hall, 'a very small, odd little black figure') are copied almost word-for-word from the real published chapter rather than newly written, so that section isn't really this writer's own prose.
The Renfield mystery and the forged-letter mystery both get tidy, spelled-out explanations a little too quickly, which lets some of the air out of the suspense before the final twist arrives.
It's competent and clean but a bit safe — the big confrontation with Alfred and the room search go by smoothly without ever really raising the emotional temperature.
Automated screening results across the ten outputs.