Why Algorithms Can’t Predict Your Love Life with Dr. Paul Eastwick

What It Covers

Dr. Paul Eastwick, author of *Bonded by Evolution*, challenges three core "Evo script" myths about human attraction — mate value hierarchies, hardwired gender differences, and short-term versus long-term partner types — presenting research showing compatibility is built through repeated interactions over time, not predicted by algorithms or personal attributes.

Key Questions Answered

• •Mate Value Myth: Attraction research using the "popularity, selectivity, compatibility" three-part model (developed by Dave Kenny) shows compatibility accounts for the largest share of attraction — even in first impressions. Only 4% of faces receive universal agreement on attractiveness ranking, meaning 96% of people receive split evaluations, making fixed desirability hierarchies statistically indefensible.
• •Gender Preference Gap: Survey-based studies show men prioritize attractiveness and women prioritize earning potential in partners. However, speed-dating "revealed preference" studies — measuring actual choices with real people — show men and women respond identically to both ambition and physical attractiveness, indicating stated preferences do not predict real-world attraction behavior.
• •Algorithm Failure: Research by Dr. Samantha Joel used machine-learning models with extensive self-reported data to predict romantic compatibility between pairs — replicating what dating apps do — and predicted nothing. Algorithms can identify individual selectivity and popularity, but cannot identify which two specific people will connect, making algorithmic matching fundamentally unreliable.
• •Compatibility as Creative Chaos: Romantic compatibility is constructed through repeated, unpredictable interactions rather than pre-existing similarity or matched preferences. Studies show agreement on attractiveness decreases as people know each other longer, meaning initial "mate value" advantages erode over weeks and months while unique pair-specific chemistry grows — a process resembling summer camp social dynamics.
• •Dating Strategy Reframe: Rather than optimizing dating profiles or filtering by deal-breakers, expanding social networks through mixed-gender friend groups produces better relationship outcomes. Heterosexual men and women with cross-gender friendships find romantic partners more reliably — not by dating those friends, but through the introductions those networks generate organically over time.