◉ PSYCHOHISTORY

In the Network of the Conclave (Soda, Iorio, Rizzo — Bocconi, May 2025)

artifact
Three business-school researchers used math to call the next Pope before the pundits did.
Who they are

A May 2025 research paper by Giuseppe Soda, Alessandro Iorio, and Leonardo Rizzo of SDA Bocconi in Italy, posted online just before the conclave that elected the new Pope.

What they do

It mapped the web of relationships among the cardinals and used that map to spot who was best positioned to win — correctly flagging Robert Francis Prevost, who became Pope Leo XIV.

How it works

They charted who served together on Vatican committees and who ordained whom as bishops, then measured each cardinal on three things: how central they were, how much they connected different factions, and how well they could build alliances. Prevost scored uniquely high on being the connective hub, something mainstream betting and punditry missed.

Why it matters

It's outside proof that who rises to the top of a closed, secretive hierarchy can be predicted just by measuring the network — without guessing anyone's motives. The engine treats this as confirmation that its own math works, not as a new theory.

The engine's record — word for word
SDA Bocconi paper (Giuseppe Soda, Alessandro Iorio, Leonardo Rizzo), published prior to the May 8, 2025 conclave on arxiv (2505.17635). Multiplex network mapping of the College of Cardinals using curial co-memberships and episcopal consecration lines; calculated three structural metrics: status (eigenvector centrality), mediation power (betweenness centrality), coalition building. Successfully identified Robert Francis Prevost as uniquely advantageous in eigenvector centrality — ignored by mainstream punditry and baseline probabilistic models. **Engine framing — methodological-grounding artifact (Layer-2 calibration, NOT Layer-1 framework promotion per Apr 23 / Apr 24 discipline):** external peer-reviewed validation that elite ascensions inside closed hierarchies are predictable from network-topology measurement (status centrality in closed-information flows) without intent-attribution. Validates the kind of math the engine already does; does NOT introduce a new framework.
Follow the trail
Walk this on the live map →
Part of the Psychohistory engine — 2,426 entities, 6,314 documented connections. Open data, built to be proven wrong.