Anti-Collusion at Micro Stakes: Typical Deal-Making Schemes and How Poker Rooms Detect Them

Micro-stakes poker is where many players learn the game, but it’s also where collusion can feel “cheap” to run and hard to prove. The good news is that modern poker rooms treat coordinated play as a core integrity issue in 2026: accounts get reviewed, funds can be frozen, and affected players may be reimbursed. What follows is a practical, player-focused guide to the most common collusion patterns at micro limits, what evidence actually matters, and how detection tends to work behind the scenes.

What collusion looks like at micro stakes (and why it’s profitable)

Collusion is any coordinated behaviour where two or more players share information or adjust decisions to gain an unfair edge against the rest of the table. At micro limits, the “edge” is often created by reducing variance for the colluders rather than making brilliant bluffs. That can be as simple as two friends refusing to play big pots against each other while applying pressure to everyone else.

The micro-stakes environment encourages this because a single account doesn’t need a huge win-rate to look “normal”. A small, steady uplift from soft-play or chip-dumping can be hidden inside the noise of beginner mistakes. The smaller the stakes, the more likely the room’s traffic is large, tables are anonymous to casual players, and the average opponent won’t run deep analysis of hand histories.

It’s also common for collusion to overlap with other integrity breaches: multi-accounting, account sharing, and “ghosting” (someone else playing your account), because these tactics complement each other. In practice, the same group might use several methods depending on whether they’re playing cash games, sit & gos, or low buy-in tournaments.

Typical schemes: soft-play, chip-dumping, and “information sharing”

Soft-play is the classic pattern: two accounts avoid value lines against each other that they routinely take versus everyone else. You’ll see odd checks back on safe rivers, unusually small bet sizes in clear value spots, or folds that don’t match how the same player behaves in similar situations against third parties. One hand proves nothing, but repeated “kindness” in the same direction starts to matter.

Chip-dumping is more direct. One account transfers chips by making calls or raises that are clearly negative EV, especially in tournament contexts where stack distribution is powerful. A common version at micro buy-ins is “late reg dumping”: one account enters, spews chips to the partner, and stops. Another is “bubble dumping”, where a short stack shoves far too wide into a partner who calls too tight or too wide in a way that conveniently protects the colluder’s tournament life.

Information sharing can be as blunt as telling a friend your hole cards in real time, but it also shows up as signalling: unusual timing patterns, repetitive bet sizing “codes”, or coordinated chat behaviour. Some teams do it off-client (messaging apps, voice calls), which makes it invisible to players but still detectable via gameplay patterns, table overlap, and metadata the room collects.

Red flags you can spot without overreacting

The biggest mistake players make is treating a single strange hand as proof. Micro-stakes are full of strange hands for honest reasons: inexperience, tilt, or people experimenting. The aim is to identify repeated, directional patterns that are hard to explain by chance, then report them with evidence rather than accusations.

Start with table overlap and frequency. If the same two screen names appear together across many sessions, sit together quickly, or repeatedly end up on the same short-handed tables, that’s a basic data point. On its own it’s not a breach—friends can coincidentally play the same hours—but it becomes relevant when combined with soft-play or chip transfer hands.

Then look for asymmetry. In normal poker ecosystems, two regulars might have a cautious dynamic, but it won’t be dramatically different from their approach against other regs. In collusion, the “special relationship” tends to show up as lopsided aggression: one player avoids putting the other to tough decisions, yet punishes everyone else relentlessly in similar spots.

Hand-history patterns that are genuinely suspicious

Repeated “no-contest” pots are one of the cleaner signals. For example, Player A opens, Player B calls, a dry flop comes, and they check down to showdown at a frequency that doesn’t match either player’s usual c-bet and turn-barrel habits. If both players are otherwise active, that contrast is meaningful.

Another pattern is the “protected squeeze”: Player A opens, one or more players call, Player B squeezes large, and Player A folds far too often compared to how they respond to squeezes from other players. If you see this happen repeatedly, it can indicate they’re coordinating to isolate weaker players and avoid fighting each other for stacks.

In tournaments, watch for unbalanced ICM behaviour between the same pair. Examples include unusual passivity when they collide (especially near the bubble), followed by extremely aggressive lines against third stacks of similar size. Again, the point isn’t that micro players misplay ICM—they do—but that the misplays consistently benefit the same partner.

How poker rooms detect collusion in 2026

Most poker rooms won’t publish their exact detection thresholds because it helps cheaters adapt. What they do communicate is the general approach: large-scale analysis of gameplay data, relationship mapping across accounts, and investigations that can lead to freezes, bans, and reimbursement to affected players. Integrity teams typically combine automated flagging with manual review.

From a technical angle, modern detection goes beyond “these two players soft-played once”. Rooms can examine long-run frequencies across millions of hands, compare behaviour to population baselines at the same stakes, and test how unusual certain sequences are statistically. If two accounts repeatedly produce patterns that are rare in honest pools—especially in the same direction—that triggers deeper review.

Crucially, the room can also use account-level and device-level information that regular players cannot see. That includes connection data, device fingerprints, location signals, and behavioural fingerprints such as consistent play schedules and session overlap. This is why a good report doesn’t need to claim you’ve “proven” anything; you’re giving the integrity team a strong starting point.

What “evidence” rooms rely on: beyond the cards

Gameplay evidence usually starts with correlation: how often two accounts share tables, how they play pots against each other, and whether their decisions are unusually aligned with protecting each other’s stack. Network analysis helps here: if several accounts orbit the same set of tables and show “friendly” dynamics with each other, that looks less like coincidence.

Metadata is the part cheaters underestimate. Even if a group avoids obvious chip-dumping hands, consistent links—same device signature, repeated logins from related networks, similar session timings, and patterns of table selection—can connect accounts. Rooms also look for attempts to mask identity, such as unusual VPN/proxy usage or remote-desktop behaviour, because those are common in organised cheating setups.

Finally, the investigation stage matters. A room may freeze suspicious funds to prevent withdrawal while reviewing a case, and if wrongdoing is confirmed, confiscated balances can be redistributed to impacted players. That’s one reason reports should be calm and factual: the integrity process is closer to an internal audit than a public argument at the table.