Most five-stacks lose because they're five individuals who happened to queue together. A real team has a plan for the draft phase, defined roles that match actual player strengths, and a shotcaller who makes the final call. The difference between a stack and a team is structure — and structure is what this guide builds.
The 5 Attack Roles (and What They Actually Do)
A balanced attack covers five jobs. Not five fraggers, not five supports — five jobs.
1. Hard breacher. Opens reinforced walls and hatches. Thermite, Ace, Hibana, Maverick. Without one, entire sites are unplayable. This is a non-negotiable slot on most sites — if your team has no hard breach, the round plan collapses before it starts.
2. Support / gadget clear. Removes defender utility so the breach and entry can work. Thatcher (EMP), Kali, Twitch, IQ, Flores. This player reads the defense and dismantles it piece by piece. The most undervalued role in ranked — good support players win rounds before a single gunfight happens.
3. Entry fragger. Takes first contact, creates space, dies for information when needed. Ash, Zofia, Sledge, Iana. The entry's job is not to survive — it's to force the defense to react, trade favorably, and open the map for the rest of the team.
4. Flank watch. Protects the team's back and holds map control. Nomad, Gridlock, Jackal. Every push has a flank route the defense will use; without someone watching it, your execute ends with two defenders shooting you in the back. The most commonly skipped role in casual stacks — and the one whose absence gets blamed on "bad luck."
5. Flex. Fills whatever the site, map, or enemy comp demands. Could be a second breacher, a second entry, intel (Dokkaebi, Lion), or a plant specialist. The flex player is your most adaptable teammate — the one who can read the round and become whatever's missing.
Note: operator metas shift with every balance patch. These examples reflect the current meta, but the *roles* don't change — only which operators fill them best.
The 5 Defense Roles (and What They Actually Do)
1. Anchor. Holds site, plays post-plant denial, survives. Rook, Doc, Maestro, Smoke, Thunderbird. Anchors win rounds in the last 30 seconds by being alive, positioned, and armed with denial utility. If your anchors die early, the site is gone.
2. Roamer. Applies pressure away from site, wastes attacker time, gathers intel. Vigil, Mozzie, Oryx, Caveira. A good roamer doesn't need kills — they need to force the attack to spend drones, time, and utility hunting them. Every minute wasted on a roamer is a minute not spent on the execute.
3. Intel gatherer. Provides eyes for the team. Valkyrie, Echo, Maestro, Pulse. Intel wins rounds in Siege more than aim does — knowing where the push is coming, where the plant is going down, where the last attacker is hiding. The intel player's value shows up in everyone else's stats, not their own.
4. Utility denial. Stops the attack's plan before it starts. Mute, Kaid, Bandit, Azami. Denial operators force attackers onto plan B — Mute jammers eat drones and breach charges, Kaid electrifies hatches, Azami reshapes the site. Denial is the defense's answer to the attacker's execute.
5. Flex. Adapts to the site and the enemy's tendencies. A second roamer against a slow team, a second anchor against a rush comp, extra intel when the attackers are unpredictable.
The critical mistake: stacking similar operators. Five roamers means nobody's on site. Five anchors means attackers own the entire map for free. Balance is the whole game.
Assigning Roles to Real Players (Strengths, Not Preferences)
This is where stacks fail. Everyone *wants* to entry frag. Nobody *wants* to play Thatcher. So the team ends up with three Ash mains and no one who can open a wall — and then wonders why they lose.
Assign roles by strengths, not preferences:
Ask three questions about each player:
1. What's their best mechanical skill? (Aim → entry/roamer. Positioning → anchor/flex. Utility usage → support/denial.)
2. What's their communication style? (Talkative and decisive → IGL/shotcaller. Quiet but observant → intel/anchor. Loud and aggressive → entry.)
3. What do they actually enjoy enough to play 50 rounds of? (A miserable support player is worse than a mediocre one — enjoyment sustains consistency.)
The strengths-first assignment looks like this: your best aimer entries, even if they'd rather anchor. Your most patient player anchors, even if they find it boring. Your smartest reader of the game plays support or IGL, because those roles multiply everyone else's effectiveness.
Handle the preference problem directly. Be honest as a group: "We need a Thatcher player. Who's willing to learn it?" Rotate the unpopular roles across sessions so no one burns out. And remember — in ranked, the support player with a 1.0 K/D who enables three teammates is more valuable than a fourth fragger.
For players still building their operator pool, start with our beginner operators guide — it covers the simplest picks for each role.
The Ban Phase Playbook
Bans are the first teamfight of every match, and most stacks waste them on whatever annoyed them last game. Run bans like a strategy:
Step 1: Know your map. Each map has 2–3 operators who define it. On maps with critical reinforced walls, banning the enemy's best hard-breach counter changes everything. On intel-heavy maps, banning Valkyrie or Echo reshapes the defense.
Step 2: Ban what beats YOUR comp, not what's "OP." If your attack plan relies on a Thatcher-Ace combo, ban the operators who shut it down (Kaid, Bandit trick counters). If your defense leans on Mute, ban the attackers who clear him effortlessly.
Step 3: Ban the enemy's comfort, not yours. If you know the enemy team (rematch, recognizable stack), ban their known mains. A team forced off its best operators plays measurably worse.
Step 4: Have a default ban set. For each map, agree on 2 attack bans and 2 defense bans in advance. Defaults prevent the 15-second panic where someone bans randomly. Adjust from defaults based on the enemy, don't invent from scratch every time.
Step 5: Communicate the why. "Banning Mira because we're attacking Bank basement" takes three seconds and aligns the whole team's expectations. Silent bans breed confusion about the plan.
Building Default Strats (2–3 Per Map)
You don't need a playbook the size of a pro team's. You need 2–3 default attacks and 2–3 default defenses per map in your rotation — simple, practiced, executable under pressure.
What a "default strat" actually is: who goes where, what utility gets used in what order, and what's the trigger to execute. That's it. Not a 47-step plan — a skeleton the team can run without thinking.
Example attack default (any bomb site):
- Hard breacher + support take the main breach wall
- Entry + flex take map control toward the secondary push
- Flank watch holds the cut-off
- On the IGL's call (or at 1:30 remaining), everyone converges
Example defense default:
- 2 anchors set up site with denial utility
- Roamer takes their position with a fallback plan
- Intel player places cameras/drones, then plays their role
- On the attack's execute, anchors play crossfires while roamer flanks
Build them in custom games, not in ranked. Spend 20 minutes per map in a custom lobby: walk the default, place the utility, time the execute. One practice session per map saves ten ranked rounds of improvisation.
Name your strats. "Default A," "rush B," "vertical take" — short names the IGL can call in three syllables. In the round, there's no time for paragraphs.
Comms Structure: Shotcaller and Callout Discipline
Five people talking at once is noise. Structure turns it into information.
Designate a shotcaller (IGL). Not a dictator — the person who makes the final call when the team disagrees. "We're executing in 10 seconds" ends debate. The IGL doesn't need the best aim; they need the clearest read of the round and the calmest voice.
Callout discipline rules:
1. Location first, always. "Two on mezzanine" beats "they're pushing!" — the team needs to know *where* before *what*.
2. Dead players give intel, then go quiet. Call what you saw, then stop talking. A dead player narrating the round is noise.
3. Confirm executes. "Executing now" → everyone pushes. No one freelances during the execute.
4. Keep it short. Three-word callouts. If it takes a sentence, it's too long for mid-round.
Pre-round comms (30 seconds): IGL states the plan, everyone confirms their role. "Default A, I'm breaching, entry takes blue, flank watch holds lobby." Five seconds of alignment prevents 30 seconds of chaos.
Post-round comms (10 seconds): one sentence on what worked or didn't. "They stacked the breach, we need the secondary push next time." Then move on — post-round is for adjustments, not arguments.
The 3 Classic Stack Conflicts (and How to Fix Them)
Conflict 1: Everyone wants to frag. Four players instalock Ash, Zofia, Sledge, and Iana. No breacher, no support, no flank watch. The round is lost in the operator select screen. *Fix:* enforce role coverage before preferences. "We need breach and support locked before anyone picks entry." Make it a rule, not a request. The players who flex to support roles are the reason the fraggers get to frag.
Conflict 2: Nobody wants to IGL. Everyone's happy to shoot, nobody wants to make calls — so nobody does, and the team plays five simultaneous solo queues. *Fix:* rotate the IGL role across sessions until someone takes to it. Frame it honestly: the IGL doesn't need to be the best player, just the most decisive. And give the IGL real authority — if the team ignores calls, the role is decorative.
Conflict 3: Role overlap and redundancy. Two roamers who both leave site. Two hard breachers on a site with one reinforced wall. Support players who both bring Thatcher. *Fix:* the operator select screen is a team meeting, not five individual decisions. Talk through picks: "I've got breach, who has support?" Overlap isn't always bad — doubling up on a key role for a specific strat is fine — but accidental overlap is just waste.
The deeper fix for all three: build your stack on trust and shared goals, not just skill. A slightly worse player who communicates, flexes roles, and doesn't tilt is worth more than a star who instalocks, goes silent, and blames everyone. Vibes are a competitive advantage.
Adapting Comp to Map and Site
Defaults are starting points, not laws. Adapt:
Map factors: vertical-heavy maps (more soft floors) reward Sledge/Buck and punish anchor-heavy defenses. Long sightline maps reward DMRs and intel. Tight maps reward shotguns, SMGs, and trap operators.
Site factors: some sites have one critical reinforced wall (breach is everything); others are about map control (flank watch and roamers matter more). Read the site, then pick the comp — not the other way around.
Enemy factors: against a rush-heavy team, add denial and traps. Against a slow default team, add roamers and intel to punish their setup time. Against a team that always bans your breacher, have a backup breach plan.
The adaptation rule: change one thing at a time. Swapping your entire comp mid-match creates confusion; adjusting one role to answer what the enemy is doing keeps everyone oriented.
When to Break the Mold
Balanced comps win most rounds. But sometimes the right play is deliberately unbalanced:
When you're much better mechanically: five fraggers can overwhelm a weaker team through pure gun skill. Don't do this against equals.
When the enemy has a pattern: if they always play the same defense, a hyper-specific counter-comp (triple denial, all intel) breaks them.
When the round demands it: match point, 1:00 left, you need the plant — pick the five operators who get the plant down, not the five who look balanced on paper.
When morale needs it: sometimes the team just needs a fun round. A meme comp after three straight losses can reset the mental state better than any timeout. Just don't make it a habit.
Breaking the mold works when it's a *choice* — a deliberate deviation from a solid default. Random comps aren't breaking the mold; they're just random.
Frequently Asked Questions
What's the ideal 5-stack composition for ranked?
One hard breacher, one support, one entry, one flank watch, one flex on attack. Two anchors, one roamer, one intel, one denial on defense. Adjust per map and site, but start here.
Who should be the IGL?
The most decisive communicator, not necessarily the best fragger. The IGL needs game sense, calm under pressure, and the team's trust. Rotate the role until the right person emerges.
How do we handle someone who won't play support?
Make role coverage a team rule, not a personal favor. Rotate unpopular roles across sessions. And be honest: a stack where one person refuses to flex is a stack with a ceiling.
Should our stack have fixed roles or rotate?
Fixed primary roles with flex capacity. Each player mains one role but can cover a second. Total rigidity breaks when someone's having an off day; total flexibility means no one masters anything.
How many strats do we need per map?
Two to three defaults per side, practiced in customs. Depth on a few strats beats a binder full of plans nobody remembers under pressure.
Do we need a coach or analyst for ranked?
No — but one player should own VOD review. Watching one replay per session and noting two adjustments is 80% of what an analyst provides at ranked level.
The Bottom Line
A stack is five players in a lobby. A team is five roles filled by the right players, a ban plan, two practiced strats per map, a shotcaller, and the discipline to flex when the team needs it. Build the structure, resolve the conflicts honestly, and your five-stack stops being five solo queues that happen to share voice chat — and starts being the reason you climb.
For the solo-vs-stack decision itself, see our solo queue vs stacking breakdown. For the individual climbing fundamentals, work through the climbing playbook. And track your team's progress with the R6 Player Tracker.