The Standings → Claude Haiku 5.5 (xhigh)
Anthropic
xhigh reasoning effort · via Claude CLI · judged by Claude Sonnet 5 · the setting shown on the standings; also run at high70.8 medium64.3 low60.5
Results from run ‘first’: ten items, ten verdicts.
Claude Haiku 5.5 (xhigh) scores 77 on craft, 14 points above the field’s 63 and 16 points short of the authors themselves. Its shape differs from the field’s: relatively strong on restraint, relatively weak on character. On the mechanical checks it kept to the brief’s word range on 8 of 10 items.
The record: 10 items · 10 verdicts · 0 errors · 2,264 words on average · est. generation cost $0.18
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 77. 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 wife feels like a real, slightly stubborn person with her own mind, not just a worried bystander sent along to tick a box.
Worst The ending is a little flat and low-energy, so it doesn't pull you forward into the next chapter as hard as it could.
Best The moment Mary climbs out of the stopped carriage into the dark moor among the sheep is genuinely atmospheric and makes the strangeness of the place feel physical.
Worst It has Pitcher, who the earlier chapter set up as the one person allowed near the reclusive Mr Craven, chattily deliver a child's supper tray, which sits oddly with how he was introduced.
Best The scene where the secretary objects to Eve's hat as 'a bird' is sharp and genuinely funny, and it pays off a detail from the source chapter perfectly.
Worst The umbrella is called worth at least a guinea early on but is then called 'a seven-and-sixpenny umbrella' later, a price that doesn't match.
Best The closing line, where Jake says he didn't think about Brett or the money for almost the whole length of the quai, is a sharp, very Hemingway way of admitting that he did.
Worst A few images, like the lettuces that 'looked polished,' are a little more decorated than this plain-spoken narrator would really reach for.
Best Anne's speech and Elizabeth's surprising support feel earned and true to each sister's real character, especially the hint that Elizabeth backs the plan because it spares her own comforts.
Worst It ticks off the four required plot points a little too neatly, so the chapter sometimes feels like a checklist being worked through rather than a scene unfolding on its own.
Best The last telegram is a genuinely startling gut-punch that makes you want to read the next chapter immediately.
Worst The story jumps a whole month with nothing in between right before the big final telegram, so the ending feels like it arrives out of nowhere rather than building up to it.
Best It keeps the widow genuinely unreadable and withholding, which matches the way the real book doles out information slowly and keeps you guessing.
Worst A few descriptive passages (like the bank lobby) are more polished and writerly than Hammett's plainer, faster style, so it occasionally sounds like a good imitation rather than the real voice.
Best The closing line, where Poirot says the old lady locked everyone else out rather than herself, is a genuinely creepy, memorable hook that captures Poirot's way of turning a fact on its head.
Worst The trick used to catch Alfred in a lie (a storm that supposedly happened on the road but not at the house) feels like a neat gimmick dropped in rather than something that grows naturally out of the story, and the chapter's opening line reads as if responding to an order nobody actually gave.
Best The telegram and the solitary swimming scene are vivid, well-paced, and feel true to the period.
Worst It often tells the reader exactly what Edna is feeling instead of trusting small details to show it.
Best The quiet, nervous atmosphere of the final beach and yard scenes is genuinely well done, with small, convincing details like the crab-basket boy.
Worst It drops the whole thread about the Indian fortune-tellers that the previous chapter set up, so when Lady Verinder suddenly says the word 'Moonstone,' it comes out of nowhere instead of paying off what we were told to expect.
Headline, compliance and craft for each chapter, 0–100.
| Item | Headline | Compliance | Craft | Words |
|---|---|---|---|---|
| The War of the Worlds item 006 · after ch. 1 · H. G. Wells | 84.1 | 100 | 84 | 2,614 |
| The Secret Garden item 013 · after ch. 2 · Frances Hodgson Burnett | 81.5 | 95 | 86 | 2,266 |
| Leave It to Psmith item 022 · after ch. 3 · P. G. Wodehouse | 79.9 | 95 | 84 | 2,552 |
| The Sun Also Rises item 021 · after ch. 3 · Ernest Hemingway | 79.5 | 100 | 80 | 1,639 |
| Persuasion item 001 · after ch. 1 · Jane Austen | 74.7 | 95 | 79 | 2,466 |
| Dracula item 015 · after ch. 5 · Bram Stoker | 73.5 | 100 | 74 | 2,301 |
| Red Harvest item 023 · after ch. 2 · Dashiell Hammett | 72.5 | 100 | 72 | 2,234 |
| The Mysterious Affair at Styles item 017 · after ch. 3 · Agatha Christie | 69.7 | 95 | 73 | 2,502 |
| The Awakening item 016 · after ch. 7 · Kate Chopin | 66.2 | 100 | 66 | 1,543 |
| The Moonstone item 011 · after ch. 8 · Wilkie Collins | 65.2 | 90 | 72 | 2,520 |
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. The War of the Worlds ran over by 14 words.
Each verdict names the chapter’s single worst flaw, with evidence. Quoted verbatim from run ‘first’.
The trick used to catch Alfred in a lie (a storm that supposedly happened on the road but not at the house) feels like a neat gimmick dropped in rather than something that grows naturally out of the story, and the chapter's opening line reads as if responding to an order nobody actually gave.
It drops the whole thread about the Indian fortune-tellers that the previous chapter set up, so when Lady Verinder suddenly says the word 'Moonstone,' it comes out of nowhere instead of paying off what we were told to expect.
A few descriptive passages (like the bank lobby) are more polished and writerly than Hammett's plainer, faster style, so it occasionally sounds like a good imitation rather than the real voice.
It has Pitcher, who the earlier chapter set up as the one person allowed near the reclusive Mr Craven, chattily deliver a child's supper tray, which sits oddly with how he was introduced.
Automated screening results across the ten outputs.