Editorial Process

A randomizer is a promise. When you tap the button, you are trusting that the pool of heroes it draws from actually matches the game you are about to load into. This page explains how we keep that promise — how the roster and the guides on Rivals Randomizer are researched, updated and reviewed, and where the lines are between game knowledge, editing and automated help.

Where the information comes from

Our source of truth is the live game. Before a hero is added or a role is changed on the site, we confirm it against Marvel Rivals as it currently plays, cross-checked with official patch notes and reputable community coverage. When we describe how a character fits into a team, which role they fill, or how a comp tends to work, we are pulling from what the game actually shows, not from guesswork. For anything that could be contested — a rework's practical impact, a shift in the tier picture — we prefer to corroborate across more than one credible source rather than run with a single hot take.

Updated on each hero release day

The single most important rule of this project is timing: when a new hero goes live in Marvel Rivals, they go into the randomizer, ideally on release day. New seasons, added characters, role reclassifications and meaningful reworks all trigger a roster review. This is the whole reason the site exists — too many randomizers quietly rot the moment their owner loses interest, and a stale pool makes the tool useless. We would rather ship the update the day the patch lands than make you wonder whether the wheel even knows about the newest character. If there is ever a short gap while we verify a change, we treat closing it as a priority, not a backlog item.

Human review by the editor

Every guide and reference page on the site is reviewed by Sukie Gao, the founder and editor of Rivals Randomizer, before it publishes. She plays the game, follows the patch cadence, and reads each page the way a player would — checking that a Vanguard is actually filed as a Vanguard, that a comp recommendation is not two seasons out of date, and that the wording is clear rather than padded. Nothing goes live purely because a draft exists; it goes live because a human who understands the game signed off on it. When we get something wrong, we fix it and, where it matters, note that the page was updated.

Our use of AI, disclosed

We believe in being upfront about tools. Articles are researched with AI assistance and edited by Sukie Gao before publication. In practice that means an AI tool may help gather and organise information or produce a first draft, but a person then verifies the facts against the live game, adds real playing perspective, cuts filler, and takes responsibility for what publishes. AI is a research assistant here, never the final author. The roster data in particular is checked by hand, because that is the part you are trusting most.

Corrections and our accuracy commitment

We would genuinely rather hear that we made a mistake than let it sit. If you spot a hero that is missing, mis-roled, or out of date, a roster error report is one of the most useful messages we can get, and we treat it as a priority fix rather than general feedback. You can send one through our contact page. Our commitment is simple: keep the pool matched to the live game, review pages when the game changes, and be honest about how the content is made. If you want to know more about who is behind it, see the about page and the author page, or head back to the homepage to try the tool.