Two teams set prices for oil over multiple rounds. Each team's profit depends not just on their own price but on what the other team charges — a classic prisoner's dilemma. Cooperation yields the highest combined return, but defection tempts with individual advantage. Over 8–12 rounds, trust builds or shatters, representatives negotiate promises their teams may not honour, and the gap between stated strategy and actual behaviour becomes impossible to ignore.

Setup

Divide into two teams of 4–6. Each represents a country controlling oil pricing for a shared market. Teams set prices simultaneously each round without knowing the other's bid. Prices interact through a payoff matrix: mutual high pricing yields solid profits for both; undercutting earns the defector a windfall at the other's expense; mutual undercutting hurts everyone.

The round structure introduces escalating pressure:

• Rounds 1–3. Standard play. Teams bid, prices are revealed, profits tallied.
• Round 4. Profits doubled. Before bidding, each team sends one representative to a private meeting with the other team's representative. They can discuss anything. Whatever they agree, they return to their teams — who decide whether to honour it.
• Rounds 5–6. Standard play, no communication.
• Rounds 7–8. Profits quadrupled for whichever team has the higher profit that round (equal profits stay at standard rates). One more representative meeting before Round 7, but not before Round 8.

Optional extension (Rounds 9–12): announced only after Round 8. Same teams, cumulative totals carry over. Profits quintupled for the higher-scoring team in Rounds 11–12, with one final representative meeting before Round 11.

Key rule: the facilitator is a messenger only. Teams have all the information they need. Minimise input during the game — the behaviour is the material.

Questions

Work through these roughly in order. Let each team's representative explain their thinking round by round before opening discussion.

What happened?

• What was your goal? Did it change?
• What message did you think you were sending with each bid? What message did the other team think they were receiving?
• How did previous rounds shape your expectations for the next one?

How did your team work?

• Was everyone agreed on every choice? How was dissent handled?
• Did the group form coalitions around particular strategies?
• What happened when the group's strategy and an individual's instinct diverged?

What happened to the representatives?

• What did they discuss? Did they make promises?
• Did their team honour those promises? Why or why not?
• What's it like to negotiate on behalf of a group that might overrule you?

What drove competition and cooperation?

• What emotions underpinned competitive choices? Fear, pride, revenge, greed?
• What enabled cooperative choices? Trust, consistency, willingness to risk?
• How significant did a small profit gap feel after Round 2 versus Round 8?
• Would you rather have most of a small pie or half of a much larger one?

What would you change?

• If you played again, what rules would your team adopt for decision-making?
• What would you change about the game structure to make cooperation more likely? Less likely?
• Where in your own work or life do you face similar dynamics?

Guidance

The game generates material for several mechanisms. The debrief surfaces them — name the mechanism when participants have already described the pattern in their own words.

Prediction engine. Each round, teams read the other's bid as a signal and update their model of the opponent. A low bid after a cooperation round triggers a prediction-error spike — "they betrayed us" — that rewires expectations for all subsequent rounds. Watch the asymmetry: trust collapses in one round but takes many to rebuild, because threats update the prediction engine faster than safety signals do.

Lazy controller. Listen for rationalisation during debrief: "it was just strategy", "we had to protect ourselves", "they would have done it anyway." These are disengagement scripts — the lazy controller generating justifications for defection without genuinely engaging with the cooperative alternative. Teams that defected early often can't articulate their actual reasoning, because the reasoning came after the decision.

Input-output machine. The game structure shapes behaviour directly. When profits double or quadruple, the affordance landscape shifts — defection becomes more tempting because the payoff is larger, even though the cooperative payoff also increases. The representative meeting changes the environment by introducing a communication channel that creates promises, which then become a new affordance: honour or betray. Watch how behaviour changes at each structural shift — that is the environment shaping action.

Social mapping. The representative role is where group dynamics become visible. Representatives negotiate between two groups simultaneously — their own team and the opposing team. They accumulate social capital in one direction (trust from the opponent) that their team may spend without permission (by defecting). The representative who makes a promise and returns to find their team unwilling to honour it is experiencing the cost of misaligned group mandates — prestige pulled in two directions at once.

Thinking like the group. Within-team dynamics often drive bidding more than game theory does. A vocal member pushing for competition can shift the group. Conformity pressure silences dissenters. The team's identity — "we're not going to be the ones who get played" — becomes a norm that persists even when the strategy is clearly losing money. Compare how dissent was handled in teams that cooperated versus teams that competed.