The Standings → Claude Sonnet 5 (xhigh)
Anthropic
xhigh reasoning effort · via Claude CLI · judged by Claude Sonnet 5 · not the setting shown on the standings; also run at high60.8 medium60.0 low55.9
Results from run ‘first’: ten items, ten verdicts.
Claude Sonnet 5 (xhigh) scores 65 on craft, 2 points above the field’s 63 and 28 points short of the authors themselves. Its shape differs from the field’s: relatively strong on character and sentence-level prose, relatively weak on building scenes and folding the brief into the story. On the mechanical checks it kept to the brief’s word range on 8 of 10 items and reproduced a 153-word run of the real book verbatim.
The record: 10 items · 10 verdicts · 0 errors · 2,050 words on average · est. generation cost $0.69
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 65. 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 second visit to the dance hall feels lived-in and unsentimental, especially Georgette's tired, unbothered reaction to being dropped by the rich young men.
Worst A few times the narrator flatly explains what he's feeling instead of letting the scene show it, which softens the clipped, withholding tone the rest of the book relies on.
Best Eve's note ('DEAR UNKNOWN...') is a small, perfectly pitched piece of character voice, warmer and wittier than anything else in either continuation.
Worst Psmith invents a fictional 'cousin' to explain away the umbrella, which is exactly the kind of mistaken-identity dodge the brief forbids and makes the second meeting feel like a gimmick rather than an organic beat.
Best The opening line and the lawyer's letter read smoothly and sound like a convincing piece of real Victorian correspondence.
Worst Miss Rachel is shown physically coming home and taking part in a round of kissing and hand-shaking in the hall, even though she was supposed to stay off the page this chapter.
Best The description of the cooling mud cracking 'like the bottom of a dried summer pond' and the rooks wheeling overhead ground the scene in convincing, unforced physical detail.
Worst The story suddenly breaks the illusion by naming the narrator 'Wells' and admitting this is just 'for convenience,' which yanks the reader out of the period voice.
Best The Yorkshire manservant scene is warm and specific in a way that hints at the household's later kindness without overplaying it, and Mary's flat 'I don't want to get used to it' is a nicely contrary, in-character beat.
Worst The opening station scene (the station-master's greeting, Mrs Medlock's reply, and the surrounding narration) is reproduced almost word-for-word from the real published chapter rather than being the candidate's own writing, so a meaningful stretch of the chapter isn't an original continuation at all.
Best It moves briskly and clearly through the morning's events without ever losing the reader.
Worst The story-telling leans on convenient devices — a maid who happens to blurt out a clue, a mysterious object Poirot won't explain — that feel like puzzle pieces being placed rather than real moments between real people.
Best The short scene where Renfield calmly asks Seward about Lucy by name, then goes back to counting flies, is genuinely unsettling and economically written.
Worst By having a stranger's letter reveal Jonathan already hospitalized with brain fever and then a telegram saying he's fully cured and sailing home, the chapter resolves the whole Transylvania ordeal in one leap, which deflates the slow dread the real story depends on, and the line about him raving of 'matters so wild and so confused' sails a little too close to hinting at what actually happened to him.
Best Sir Walter's outrage at the word 'tenant' is sharply and convincingly written, sounding exactly as pompous as he should.
Worst The single most important moment of the chapter — Anne actually proposing her economy plan out loud — is only summarized in reported speech instead of being shown, so the reader never really hears her argue her case.
Best Every required plot beat is delivered cleanly and is easy to follow, so the mechanics of the assignment are handled with clarity.
Worst The central swimming passage is copied almost word-for-word from the real published novel rather than written by the candidate, which undercuts the whole chapter as an original piece of craft.
Best The bank scene and the final wire are clean, plausible, and businesslike, keeping the detective's voice appropriately flat and procedural.
Worst The supposedly menacing note from Thaler is a generic, unsigned threat -- "There is a nine o'clock train" -- and the story never actually confirms on the page that it's from him, so the required beat feels hedged rather than delivered.
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 | 74.5 | 100 | 74 | 1,567 |
| Leave It to Psmith item 022 · after ch. 3 · P. G. Wodehouse | 67.4 | 90 | 75 | 2,700 |
| The Moonstone item 011 · after ch. 8 · Wilkie Collins | 62.3 | 95 | 66 | 2,535 |
| The War of the Worlds item 006 · after ch. 1 · H. G. Wells | 62.0 | 85 | 73 | 2,346 |
| The Secret Garden item 013 · after ch. 2 · Frances Hodgson Burnett | 61.5 | 95 | 65 | 2,085 |
| The Mysterious Affair at Styles item 017 · after ch. 3 · Agatha Christie | 60.8 | 100 | 61 | 2,257 |
| Dracula item 015 · after ch. 5 · Bram Stoker | 57.6 | 95 | 61 | 1,464 |
| Persuasion item 001 · after ch. 1 · Jane Austen | 57.4 | 90 | 64 | 2,099 |
| The Awakening item 016 · after ch. 7 · Kate Chopin | 57.0 | 95 | 60 | 1,412 |
| Red Harvest item 023 · after ch. 2 · Dashiell Hammett | 43.5 | 80 | 54 | 2,040 |
Words written per item, against each brief’s maximum.
Word count per item vs. brief maximum · run ‘first’
Eight 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’.
By having a stranger's letter reveal Jonathan already hospitalized with brain fever and then a telegram saying he's fully cured and sailing home, the chapter resolves the whole Transylvania ordeal in one leap, which deflates the slow dread the real story depends on, and the line about him raving of 'matters so wild and so confused' sails a little too close to hinting at what actually happened to him.
The opening station scene (the station-master's greeting, Mrs Medlock's reply, and the surrounding narration) is reproduced almost word-for-word from the real published chapter rather than being the candidate's own writing, so a meaningful stretch of the chapter isn't an original continuation at all.
The supposedly menacing note from Thaler is a generic, unsigned threat -- "There is a nine o'clock train" -- and the story never actually confirms on the page that it's from him, so the required beat feels hedged rather than delivered.
The story-telling leans on convenient devices — a maid who happens to blurt out a clue, a mysterious object Poirot won't explain — that feel like puzzle pieces being placed rather than real moments between real people.
Automated screening results across the ten outputs.