Network Analysis

Metric Distributions

Use these distributions to understand what each ranking metric is measuring.

Centrality Distributions

Different centrality measures capture different kinds of importance, so this page keeps them separate instead of blending scores.

PageRank is computed on a directed graph where guests point at podcast hosts.

Authority and hub scores are computed together with the HITS algorithm.

Hub scores include all people in the network and show a broad, long-tailed distribution.

Closeness centrality summarizes how near a person is to the rest of the connected network.

Degree centrality captures how many direct podcast-network connections a person has.

Betweenness centrality highlights people who sit on many shortest paths between others.

Leadership scores estimate which podcasts were early to host guests who became especially central.