Guides · Marvel Rivals data

Marvel Rivals map and hero matchups from 2,832 Competitive matches

Does a hero do better on some maps? Does one hero really counter another? We checked both questions against 2,832 full Competitive matches recorded by Lunar between December 2025 and November 2026. Players clearly change their picks from map to map. Whether those choices win more often is much harder to show.

How the numbers were counted

We used the same matches as our hero win rates article: Competitive, six against six, nobody leaving early, no AI players and a clear winner. Each player counts once, for the hero they spent the most time on. No player names or IDs are used.

• Picked: the share of team lineups that included the hero. Every match has two lineups.
• Against: the hero was on the other team. A matchup here is a team result, not a one-on-one duel. "Magneto against Thor" means the team with Magneto played the team with Thor.
• Expected: how often the first hero's team would win if the matchup made no difference, given how often each hero wins in general. A strong hero is expected to beat a weak one, so it only counts as a counter if it does even better than that.

Games where the same hero was on both teams are left out of the matchup figures, because one copy always wins and the other always loses.

Maps: what changes is the picks

The four most picked heroes overall (Invisible Woman, Magneto, Doctor Strange and Cloak & Dagger) are the most picked hero on every map. Further down the list, though, picks move noticeably from map to map. The table shows, for each map, the hero picked most and the hero whose pick rate rose most compared with all maps together.

The "picked more than usual" column only includes heroes with at least 40 lineups on that map.

A few shifts are large:

• Jubilee on Thebes: in 28% of lineups, against 17% across all maps.
• Cloak & Dagger on Yggdrasill Path: in 40% of lineups, against 30%.
• Magneto on Midtown: in 40% of lineups, against 33%.
• Winter Soldier on Lower Manhattan: in 30% of lineups, against 24%.

Fights also run at a similar pace everywhere. Most maps sit close to 10 KOs per team per minute, with Thebes the busiest at 10.9. Match length varies far more than pace: see what decides a Competitive match for how long each map takes.

Do some heroes win more on certain maps?

Not in a way this data can show. We checked 210 hero and map combinations with at least 50 lineups each. If maps made no difference at all, luck alone would make about 10 of them look unusually good or bad. We found 4.

In other words, a hero's win rate on one map is about what you'd expect from its win rate everywhere. You'll see claims like "Rocket Raccoon wins 68% on Arakko". With 65 games behind it, a figure like that is mostly noise. Players clearly believe some heroes suit some maps, which is why picks shift, but that doesn't turn into extra wins that we can measure.

Hero against hero: are counters real?

We looked at every pair of heroes that met on opposite teams at least 80 times: 174 pairs. For each one we compared how often the first hero's team won with what we'd expect from the two heroes' usual win rates.

If counters didn't exist at all, luck alone would leave about 8 of the 174 pairs looking lopsided. We found 5, and none stood far enough out to rule out chance. So in this data, once you account for how strong each hero is in general, no matchup clearly beats chance.

These were the ten pairs furthest above expected. Treat them as matchups to watch, not proven counters:

Pairs are ordered by how unlikely the gap is to be luck, which depends on the number of games as well as the size of the gap. That's why Emma Frost against Doctor Strange, with 254 games, makes the list with a smaller difference. Every matchup works both ways: Thor's team beat Magneto's in 43.1% of those 102 games.

Rocket Raccoon isn't a counter, he's just strong

Rocket Raccoon has the highest raw matchup numbers in the data. His team beat Emma Frost's in 71.9% of 135 games, Deadpool (Strategist)'s in 60.7% of 247 and Cloak & Dagger's in 60.4% of 159. Against all twelve heroes he met at least 100 times, his team won between about 55% and 72%.

That's what a strong hero looks like, not a counter. Rocket Raccoon wins about 55% of all his games, so most of those matchups are close to what you'd expect anyway. Only Emma Frost stands out, at about 10 points above expected, and even that is within what chance can produce across 174 pairs.

The lesson goes beyond Rocket Raccoon: a "counter" list built from raw win rates mostly shows which heroes are strong overall. Comparing against what's expected is what separates the two.

What to take from this

• Pick for the map with care, not certainty. Shifts such as Cloak & Dagger on Yggdrasill Path show what experienced players reach for, and that's useful to know, but the data doesn't show those picks winning more.
• Don't swap just because a counter chart says so. Average hero counters are much weaker than they're often made out to be. A swap to a hero you play badly costs more than a typical matchup is worth.
• Your own matchups are a different story. You may struggle against a particular hero even if the average player doesn't. Lunar's ban planner uses the enemy heroes your team actually loses to, and duo compatibility shows your best maps and enemy matchups as a pair.

Limits of this data

A team result says nothing about who won the fights between two particular heroes, and heroes are picked in response to each other, which muddies the picture. 80 to 250 games per pair is enough to spot big effects but not small ones, so modest counters of a few percentage points may well exist without showing up here. The matches come from Lunar's users and their lobbies, not a random sample of all players, and they span many balance patches. These are aggregates of real matches, not official statistics from the game's developer.

Related reading: hero win rates and pick rates and team composition in 2,832 Competitive matches.