The roster answers questions you can ask a table: who, when, how long. These are the questions it cannot, worked out from the same data file the rest of the site is built from.
Every figure quoted in them is computed from the cast list rather than repeated from somewhere else, which is the rule the whole site runs on and the reason these say things the source article does not. Where a number is arguable, the piece says so and shows the working. If you want the underlying facts instead, they are in the sortable roster, the history of the show, the biographies of all 174 cast members, and the page of statistics and a quiz.
Twenty-four anchors in fifty-one seasons, and why most of them were writers before they were performers.
Fifty-nine people lasted one season
A third of the roster appeared in a single season, and almost half of those arrived in one decade.
Featured, repertory, and the promotion that never came
Median tenure for the never-promoted is one season. For the promoted it is seven.
Four people in fifty-two seasons have a second stint, and all four left in the same collapse.
One in five alumni come back to host
Thirty-nine of 174 returned as host, and the median host stayed three times as long as the median everyone-else.
The writers who stepped in front of the camera
Fifty-eight cast members had a writing credit too, and the type has thinned everywhere, most slowly at the Update desk.
Fifteen people joined in 1980 and twelve of them lasted one season. The near-death year, in figures.
The eleven who stayed ten years or more
Nine of the eleven longest-serving cast members were hired this century. Long tenure is a modern invention.
From nine to twenty-plus, and the difference between the 1980 spike and the 2022 one.
Fifteen hires in one year, none at all in eight others, and why the cast empties out all at once.
Eleven people joined in 1985, or twelve, depending on a definition. Why counting cast members is hard.
Where the data comes from, what a season means here, and the edge cases that do not resolve cleanly.