A race swap, in WhatSphere’s data model, is a single specific event: a character whose race or ethnicity on screen differs from the race or ethnicity that character held in the source material an adaptation is drawn from. It is recorded only for adaptations, only at the level of an individual named character, and only when the source establishes a race to depart from. Original characters, ambiguous source material, and brand-new roles are never counted. Across the catalog the model currently identifies 64 such changes spread over 26 titles.
That is a deliberately narrow definition, and the resulting count is small. The point of tracking it is not to assemble a tally for its own sake but to let a reader see the actual shape of a phenomenon that is discussed far more often than it is measured — which directions of change occur, how often, and in which kinds of roles.
How WhatSphere measures this
Each adaptation in the catalog is matched against its source, and individual roles are compared character by character. When a character’s established source race differs from the casting on screen, the change is logged with its origin race, its destination race, the billing tier of the role (lead, supporting, ensemble, or cameo), and a cultural-significance rating that records whether the character’s race was incidental to the story or load-bearing for it. Every figure here is generated by WhatSphere’s scoring model, not by a critic’s judgement. Scores describe who a title appears to be made for and how its content is composed; they are not ratings of quality, and a higher or lower number is never a verdict on whether a title is good.
What the data shows
The changes are not evenly distributed across directions. The most common single direction is White to Black, accounting for 30 of the 64 recorded changes. The remaining changes are scattered thinly across 11 other directions, most of them appearing only once or twice. Because the totals are small, a difference of a few characters between directions should be read as a difference of a few characters, not as a structural trend.
Billing tier is where the data is most informative. Of the changes recorded, 20 occur in lead roles and the remainder sit in supporting, ensemble, or cameo parts. A change to a lead is far more visible to an audience than a change to a background role, so separating the two keeps a single recast supporting player from being discussed as though it reshaped a film. The cultural-significance rating does the same job from the other side, marking whether a character’s race mattered to the plot at all.
Caveats and limitations
The defining limitation here is sample size. 64 changes across the whole catalog is a small number, and it depends entirely on which adaptations have been researched and how confidently a source race could be established. Some sources leave a character’s race genuinely undefined, in which case no swap is recorded in either direction — not because nothing changed but because there was no fixed starting point to measure against. The counts will move as more adaptations are added, so any single figure should be treated as a current reading rather than a settled fact.
Why it matters
Casting changes between a book, comic, or game and its screen version are among the most heavily debated topics in entertainment, and almost none of that debate is anchored to a count. A neutral, role-weighted record of what actually changed — in which direction, in which tier, and how central the character was — replaces the impression that the practice is either ubiquitous or vanishingly rare with a number a reader can check. The titles linked below are the adaptations the catalog has analysed most closely; each one’s page breaks down its casting against its source.