Blog

I recently got into the US version of the reality show Survivor, mainly because of another TV show The White Lotus, which is written by Mike White, a contestant in Survivor: David vs. Goliath. The two shows share similar motifs - both throw a cast of strangers, fictional or otherwise, into an exotic locale where the shows focus on how different personalities interact and clash.

As a newbie to the series, I had to decide which seasons to watch first. This proved to be difficult because (as of January 2023) there are 43 seasons to choose from. To further complicate things, contestants in older Survivor seasons may return to newer seasons. A recent example is season 40 Survivor: Winners at War, where all the contestants were Sole Survivors from previous seasons. Watching this season as my first entry into the series would be a bad idea since it would spoil the outcomes of 20 prior seasons.

I used this Ringer article to figure out which seasons I should watch first, and while I found it helpful, to truly have spoiler-free viewing experience I figured that I would need to create a directed acyclic graph. And so I did:

(Click to see full image)

Each node in the graph is a Survivor season. An edge indicates that the source node season has at least 1 contestant returning to the sink node season.

Using this graph, I can see that if I want to watch Heroes vs. Villains, I should probably watch Micronesia first. Of course, to truly appreciate Micronesia, one must first view Pearl Islands, Cook Islands and China and so on to both get the backstory of the returning contestants and avoid spoilers. It’ll take a long time to get through these many seasons, but at least now I know the viewing orders.

The graph can be too complex for some seasons, so here is a table showing which seasons had returnees:

The data comes from the survivoR package. Analysis is done using NetworkX and graphviz in Python. Code