Distribution of reverts per user

SQL

SELECT
  DISTINCT revert_factor,
  CUME_DIST() OVER (
    ORDER BY
      revert_factor
  ) AS cume_dist
FROM
  (
    SELECT
      rev_actor,
      ROUND(COUNT(ctd_id) / COUNT(*), 1) AS revert_factor
    FROM
      revision_userindex AS revision
      INNER JOIN change_tag ON revision.rev_id = change_tag.ct_rev_id
      INNER JOIN actor_revision AS actor ON revision.rev_actor = actor.actor_id
      LEFT OUTER JOIN change_tag_def ON (
        change_tag.ct_tag_id = change_tag_def.ctd_id
        AND change_tag_def.ctd_name = 'mw-reverted'
      )
    WHERE
      TRUE
      AND actor_user IS NOT NULL
      AND rev_timestamp >= '20220101000000'
    GROUP BY
      rev_actor
  ) AS q
SQL

Checking query status...
