Marvel Rivals tier list explained: ranking heroes by win rate

Marvel Rivals tier lists move fast. The ranking you see on opening day of a new season can look nothing like week five. This post walks through how Counterwatch's MR tier list is actually computed: what shrunk win rate means, how team-ups change the reading, and why the per-rank filter matters more here than it does in Overwatch.

---

What the tiers actually measure

The Marvel Rivals tier list sorts every hero S through F by their shrunk win rate, computed across community-tracked matches at your chosen mode, rank, and game type. Not raw win rate. Shrunk.

MR's three roles are Vanguard, Duelist, and Strategist. A hero's placement is relative to their role: an S-tier Vanguard is beating the 50% baseline by a wide margin with enough matches to trust the number, not "the strongest hero in the game outright."

Middle tiers are where most of the roster sits. That is a feature of a roster this wide: you rarely find a hero who cannot hit around 50% in the right hands. Most of what you will gain from the tier list lives at the edges: what to avoid, and which handful are currently over-performing.

What Bayesian shrinkage actually does

Raw win rates break on small samples. A hero with ten matches at 100% raw wins looks unbeatable until the eleventh match happens. Shrinkage fixes the problem by blending each hero's raw win rate with a neutral 50% baseline, weighted by how many matches feed the sample.

Concretely, 400 imaginary 50/50 matches get added to every hero's record before the displayed win rate is computed:

• 10 games at 100% raw → (10 × 1.0 + 400 × 0.5) / (10 + 400) = 51.2%. Barely above the baseline. Fluke-proof.
• 8,000 games at 54% raw → (8,000 × 0.54 + 400 × 0.5) / 8,400 = 53.8%. Barely moves. Real signal survives.

A hero sitting at 60% across 20 matches reads identically to a hero at 50.9% after shrinkage runs. To land S-tier, you need both a strong raw win rate and enough community data to trust it.

The team-up wrinkle

Marvel Rivals has a mechanic Overwatch does not: team-ups. When specific pairs or trios of heroes share a lineup, they unlock passive or active bonuses that change how the heroes play. A hero with their team-up active is effectively a different hero: sometimes dramatically so: than the same hero on an unsupported lineup.

The tier list aggregates across every lineup state. That is the right call for the "how strong is this hero overall" question, but it hides the team-up signal. If you are deciding whether to main someone, the hero page's synergies tab shows which partners spike the hero's win rate and which pull it down.

Put together: the tier list answers "who is strong right now in general." The synergies tab answers "who should I be queuing this hero with."

Sample sizes and MR's season resets

Ranked resets hit harder in Marvel Rivals than in Overwatch. Seasons are shorter, reset cadence is aggressive, and new heroes launch every season. Early-season tier lists are noisier than mid-season or end-of-season lists because the sample is still building and a freshly-released hero has almost no match history.

Shrinkage protects you here. A hero released mid-season with a hot streak in 60 matches gets pulled back toward 50% until the sample grows. By around week four of a season, most heroes have enough data that their shrunk win rate is within a couple of points of their raw number.

Every hero's detail page shows the underlying match count. Check it before taking a placement at face value. Tens of thousands of matches means confident. A few hundred means provisional, especially if the hero is newly released or returning from a rework.

Per-rank filtering: a bigger lever in MR than in OW

Marvel Rivals' meta at Bronze is barely recognizable next to its meta at Eternity. Execution and coordination ceilings are high across the roster, so heroes that demand precise mechanics or tight positioning look weak at low ranks and dominant at high ones. The inverse also holds for heroes that thrive in chaotic uncoordinated lobbies.

The rank filter on the full MR tier list goes from "All" through each individual division: Bronze, Silver, Gold, Platinum, Diamond, Grandmaster, Celestial, Eternity. Plus a composite "Celestial+" that combines the top two tiers.

The "All" view averages across every skill level. It is fine for a macro meta check, but if you are climbing a specific rank, filter to that rank. Bronze-to-Gold gameplay looks fundamentally different from Diamond+ gameplay, and your tier list should reflect the one you actually play.

How to actually use a tier list while climbing

A tier list is a starting point, not a prescription.

Every row is a snapshot of aggregate win rates across every map, every lineup, every team-up combination. It does not know what your team locked, what the enemy team locked, or which hero you have 300 hours on. Those three factors drive your actual win rate more than any one-tier delta does.

Concrete recommendations:

• Do not swap off mastery for a tier bump. A mastered B-tier hero beats an unfamiliar S-tier hero for most players.
• Use the hero page for decisions. Click any tier list row to see that hero's counters, synergies, and per-map numbers. That is where the real mid-queue decisions happen.
• Cross-reference with best one tricks. The tier list answers "what is strong." The best one tricks page answers "which hero rewards mastery." For a climbing main, the second question usually matters more.
• Use the Counterwatch app mid-match. The in-game overlay runs during your match, surfacing hero scores as the enemy team locks picks: no alt-tab, no dig-through-filters.

For the long-form on how we collect, shrink, refresh, and display these numbers, see the methodology page.