Malcolm Gladwell: Predicting Hits in Hollywood and Music with Algorithms
The New YorkerJuly 22, 201458 min132,272 views
27 connections·40 entities in this video→The Quest for Predictable Hits
- 💡 Malcolm Gladwell introduces Dick CoPegan, a lawyer obsessed with movies, who was deeply moved by the film "Dear Frankie," leading him to question why certain stories resonate so powerfully.
- 🧠 CoPegan's emotional reaction to "Dear Frankie" sparked a desire to understand if there was a formula for engineering emotional responses in audiences, contrasting with the common belief that art is purely subjective.
Deconstructing Emotional Impact
- 🎯 Gladwell uses the death of Princess Diana as an example, arguing that its profound emotional impact stemmed from a confluence of specific, emotionally charged details (e.g., lover, tunnel, Paris), not just a general tragic narrative.
- ⚠️ This idea challenges the long-held Hollywood dictum, famously stated by William Goldman, that "nobody knows anything" about what makes a movie successful, often describing it as an "a-tech market."
Algorithmic Prediction in Music
- 🎶 Mike McCready's company, Platinum Blue, developed software that analyzes the mathematical relationships of a song's structural components (melody, harmony, rhythm) to predict its likelihood of becoming a Billboard Top 30 hit with 80% accuracy.
- 📈 This system identifies 60 distinct "hit clusters" that successful songs fall into, demonstrating that even in music, there are underlying, predictable mathematical principles.
Epico Jaques: Engineering Movie Success
- 🚀 Inspired by McCready, Dick CoPegan co-founded Epico Jaques, a company that uses neural networks to analyze screenplays by reducing them to numerical scores of narrative elements, provided by "Mr. Pink" and "Mr. Brown."
- 💻 The neural network learns by iteratively weighting variables in screenplays against actual box office results, effectively creating a formula to predict a movie's commercial success based solely on its script, handling numerous variables without human bias.
- ✅ Early experiments showed remarkable accuracy, with the system correctly predicting TV pilot ratings and movie box office (e.g., The Interpreter) often within a few million dollars, using only the screenplay.
The Interpreter Case Study
- 🎬 Epico Jaques analyzed the film The Interpreter, identifying specific elements that hindered its success, such as starting in Africa (which "Americans' eyes glaze over") and the lack of locale as a strong character.
- 💰 By proposing changes like enhancing the "woman in peril" aspect and making the security guard character black, the system projected the movie's potential box office could increase from $33 million to $200 million, demonstrating the formula's commercial power.
The Paradox of the Formula
- 🎭 While powerful for predicting commercial success, the formula can restrict artistic freedom, forcing filmmakers to make choices based on predicted revenue rather than creative vision, making moviemaking "harder."
- 🎯 The system is most effective for "standard Hollywood drama or comedy" (e.g., boy-meets-girl plots) that are already somewhat formulaic, helping to optimize their commercial appeal rather than predicting or creating truly innovative outliers.
- 🌟 Gladwell notes that star power has a limited impact on domestic box office, primarily serving internal studio politics and mitigating financial bombs, rather than being a primary driver of success.
Knowledge graph40 entities · 27 connections
How they connect
An interactive map of every person, idea, and reference from this conversation. Hover to trace connections, click to explore.
Hover · drag to explore
40 entities
Chapters15 moments
Key Moments
Transcript214 segments
Full Transcript
Topics15 themes
What’s Discussed
Malcolm GladwellDick CoPeganMovie IndustryMusic IndustryPredictive AlgorithmsNeural NetworksScreenplay AnalysisBox Office PredictionHollywood EconomicsArtistic FreedomThe Interpreter (movie)Platinum Blue (company)William GoldmanStar PowerNarrative Elements
Smart Objects40 · 27 links
Medias· 10
People· 17
Companies· 4
Locations· 4
Concepts· 3
Products· 2