Marvel Rivals counters: who beats who in every matchup

Counter-picking in Marvel Rivals is messier than in most hero shooters, because team-ups, a new hero or two every season, and loose role rules all bend the matchup picture. The counter data cuts through some of that, as long as you know what the numbers are and are not telling you.

What counts as a counter

A counter is a hero whose community win rate against a specific opponent sits meaningfully above 50%, measured from real matches played by opted-in Counterwatch users, not from what the internet says. Every number on the counters page is shrunk, not raw, so a hero showing 58% over 80 matches gets pulled back toward neutral and only a matchup that holds at, say, 53% over thousands of games counts as real signal.

Rough thresholds after shrinkage: 52 to 53% is a meaningful edge, worth preferring if both heroes feel equally comfortable; 53 to 55% is a strong counter, worth swapping for even if you are less practiced on it; above 55% is rare in the aggregate, and usually rides a small sample, so check the match count before you celebrate.

Marvel Rivals edges cluster tighter than Overwatch's, because the 6v6 format dilutes any single matchup. You are up against six enemy heroes and standing alongside five of your own, so a 53% edge into one opponent still moves the needle, just less dramatically than the same edge would in a 5V5 game.

Team-ups move the number

This is the thing Overwatch does not have and Marvel Rivals players have to internalise. Team-ups change counter numbers. When a hero is on a lineup with their team-up partner active, their win rate against specific opponents shifts, and the counters page aggregates across every team-up state, so the number you see is a weighted average of active and inactive matches. If you are hunting a hero's maximum edge, check their synergies tab, and if one of the partners that spikes their win rate is already on your team, their real counter edge is higher than the aggregate shows.

Five matchups from recent data

Real numbers from recent Marvel Rivals Standard data, all ranks, all game types, and all liable to move as patches and team-up reshuffles land. Picked for role diversity and for matchups that teach something.

Black Panther into The Thing, 53.0% over ~12,400 matches

A Duelist beating a Vanguard. Black Panther's dash-reset kit eats stationary Vanguards alive, since every elimination refreshes his mobility and The Thing's positional commitments hand him clean dive windows. The Thing's crowd control is real, but by the time he lands it, Panther has usually tagged a squishy and reset out. One of the larger-sample counter edges in the current data, which makes it a confident read.

Iron Man into Hela, 52.9% over ~12,800 matches

A Duelist mirror. Iron Man's flight lets him dictate sightlines against Hela's precision kit, forcing her to shoot up at a moving target instead of picking off grounded heroes from her comfort zone. Hela out-damages him in a clean peek, but he rarely gives her a clean peek, and the mobility gap does the work.

Groot into Rocket Raccoon, 52.9% over ~12,500 matches

A Vanguard pressuring a Strategist. Groot's walls carve the fight in two and cut Rocket off from his team, and Groot's burst closes the kill before Rocket's self-sustain catches up. Rocket heals plenty in a settled fight, but he has no answer to a wall that forces a 1v1 in his backline. Large enough sample to be one of the more reliable Vanguard-into-Strategist reads in the set.

Venom into Wolverine, 53.7% over ~6,300 matches

A Vanguard countering a Duelist, the reverse of the usual story. Venom's mobility and lifesteal mean Wolverine's burst combo rarely closes, since Venom escapes the window before the kill confirms and re-engages on a fresh health pool. Wolverine wants to commit on a Venom because he is a Vanguard, but Venom's escape tools are wider than most Vanguards', and the dive turns into chip rather than a kill.

Mantis into Human Torch, 53.4% over ~1,600 matches

A Strategist counter. Mantis's sleep is one of the best single-target lockdowns in the game, and Human Torch's flight-based damage means he is often the one enemy Duelist who cannot escape a sleep without a cooldown. Good timing turns Torch from an uncatchable nuisance into a free kill. The sample here is smaller than the others, so expect this edge to move as more matches land.

Common counter-pick mistakes

Reading the number is half of it. Knowing when not to act on it is the other half. Swapping on a hunch instead of the data is a habit, not a read, so check the number before you lock. Swapping to a counter can break a team-up you were already unlocking, and sometimes holding your pick is worth more than the counter nudge. Enemy comps swap too, so lock your counter after they commit, not during selection. And maps matter. Some Marvel Rivals maps favour flight heavily and others punish it, so check the hero's per-map splits before you commit.

Before the match and during it

Before you queue, the team builder scores a full lineup, folding counters, synergies, and team-up unlocks into one win-chance number, which is handy for a stack locking a comp. Mid-match, the Counterwatch app overlay surfaces counter candidates as the enemy locks picks, so you are not alt-tabbing to the website from the spawn room. A 53% edge looks small on its own, but stack three or four across a lineup, keep your team-ups intact while you do it, and the math tilts the fight before anyone presses a button.