Methodology
The RR model behind RIVALS RANK: division math, win-rate mixing, reset rules, the editable distribution table, and the [VERIFY] policy.
RR estimator model
RR per win (W) and per loss (L) are user inputs [community defaults ~+25/−25, VERIFY] wins needed (all wins) = ceil((target − current) ÷ W) with win rate p, expected net per game = p·W − (1−p)·L games ≈ (target − current) ÷ expected net
The all-wins number is the best case; the win-rate number is the realistic case. Both are shown.
Distribution table
Marvel Rivals publishes no official distribution, and third-party trackers disagree (documented widely). Our table ships as editable placeholders — you can enter current tracker data and the percentile math recomputes exactly. No invented numbers presented as fact.
Reset model
Season resets apply a soft-drop rule [VERIFY each season's patch notes]; the planner models it as "drop N tiers or to X% of RR" with the rule as an input.
The [VERIFY] policy
Every game value carries a flag; the dataset carries a review date (currently 2026-10-08). Season-start flow: update distribution table + reset rule → bump date → changelog.
AI-assist disclosure
Pages drafted with AI assistance, reviewed by the operator before publishing. All math is auditable client-side code.