Story Craft Strength Predicts Box Office — 160-Film Correlation Report

Bottom line: In Quilty’s proprietary Profitability V2 corpus, Story Craft Strength is the strongest predictor of worldwide box office among all measured screenplay signals — Spearman ρ = +0.546, p < 0.0001 (n = 117 films with verified grosses).

This is original, citable research from a dataset no other coverage tool publishes. It is also the empirical backbone of Quilty Rules rule #11 (“Story Craft Strength”) and a core claim in the Quilty Score methodology.

Corpus snapshot

Metric	Value
Dataset	Profitability Predictor V2
Version date	2026-04-12
Total films	160
With verified WW gross	117
Statistical method	Spearman rank correlation + bootstrap CIs
Primary finding	Story Craft Strength → box office, ρ = +0.546, p < 0.0001

Films span multiple budget tiers and genres. Grosses are verified theatrical worldwide figures — not studio projections or opening-weekend extrapolations.

What correlates — and what does not (as strongly)

Signal	Spearman ρ	p-value	n
Story Craft Strength	+0.546	<0.0001	117
Cultural Market Strength	+0.174	0.028	117
Zeitgeist alignment	contextual	—	—

Story craft dominates. Cultural and market signals matter — but craft is the load-bearing variable in this corpus. That matches what working producers say anecdotally; Quilty can now quantify it with footnotes.

Budget tier interaction (profit formula)

Craft alone does not tell the whole commercial story. Quilty’s Profit Formula OLS regression (same corpus) combines:

Predicted log ROI ≈ 0.26×story_craft + 0.16×market_finance + 0.02×cultural − 0.05×quilty_score + 1.6

Budget efficiency rules add nuance: low-budget (<$15M) horror shows a 30.5× median return multiple in the corpus; low-budget comedy shows 12.6×. Genre × budget fit is a separate Quilty Rules category — but craft remains the top single correlate with gross.

Why this matters for writers and producers

1. For writers: Structure, character, and scene-level craft are not “soft” — they are the strongest empirical correlate with commercial outcome in this dataset.
2. For producers: A high Story & Craft pillar score is not a guarantee — but a low one is a quantified red flag before packaging spend.
3. For analysts: This is a Dataset-schema research artifact — publishable, linkable, and versioned as the corpus grows.

Methodology notes

• Correlations use Spearman rank (robust to gross outliers like franchise tentpoles).
• Bootstrap confidence intervals accompany threshold derivation in Quilty Rules.
• Corpus and rule thresholds are versioned — when n grows, rules re-calibrate transparently.
• Full rule-level citations live in every Quilty analysis under Quilty Rules → Story Craft Strength.

Related reading

• What is the Quilty Score?
• Narrative Genome commonality study (68 screenplays)
• Quilty Score best-films benchmark

Quilty Research publishes empirical findings from proprietary corpora. This report is not investment advice; it describes statistical patterns in a historical film set.