The Lived Experience Project
Got work?

TLEP / The rules of the game

The rules of
the game.

One simple game explains why some people keep losing, and what actually works. No maths needed. We will show you how it works, step by step, then you can play it yourself.

Start here Watch the film (27 min, YouTube), opens in a new tab
  1. The game
  2. Again and again
  3. The experiment
  4. How Tit for Tat works
  5. Why responding matters
  6. Mistakes
  7. Your turn
  8. Where do you see this?
  9. The town
  10. The numbers

01 / The game

Two people.
One choice.

Every round, two people each make one choice at the same time, without knowing what the other will do. They can cooperate: play fair, so you both do well. Or they can take: grab more for yourself, at the other person’s cost.

Here is how the points work. There are only four things that can happen.

  1. You both cooperate.

    YouC+3ThemC+3

    3 points each. A good result for both of you.

  2. You cooperate. They take.

    YouC+0ThemT+5

    They get 5. You get nothing. You played fair, and you lost out.

  3. You take. They cooperate.

    YouT+5ThemC+0

    You get 5. They get nothing.

  4. You both take.

    YouT+1ThemT+1

    1 point each. You both lose out: you could have had 3 each.

So what should you do?

If you only ever play once, taking always gets you more, whatever the other person does. So two clever people who meet only once both take, and both end up with 1 point instead of 3.

02 / Again and again

But life is not one round.

You meet the same people again and again: at home, at school, at work, on your street. What you do today, they remember tomorrow. Watch what happens when you take from someone you will see again. They play Copycat: they start fair, then do whatever you did last round.

Always taking against a copycat for six rounds
Round123456Total
YouAlways takeTake, +5Take, +1Take, +1Take, +1Take, +1Take, +110
ThemCopycatCooperate, +0Take, +1Take, +1Take, +1Take, +1Take, +15

  1. Round 1. You take. They played fair, so you get 5 and they get 0. Taking looks clever.
  2. Round 2. They remember. Now they take too. 1 point each.
  3. Round 3. Still taking. 1 each.
  4. Round 4. 1 each.
  5. Round 5. 1 each.
  6. Round 6. You finish on 10. If you had both played fair every round, you would each have 18.

CCooperate: play fairTTake: grab more for yourself+3Points that round

So the real question is: what is the best way to play when you keep meeting the same people?

03 / The experiment

1980: who plays best?

Robert Axelrod, a scientist who studied how people and countries get along, set up a competition. He asked experts in maths, psychology, economics and politics to send him computer programs. Each program was a strategy: a set of rules for when to cooperate and when to take.

Every program played every other program, 200 rounds each. Here are some of the kinds of strategy that entered, and what each does when the other player takes once.

How each strategy reacts when the other player takes once, in round 3
Strategy123456
The other playerTakes once, in round 3CooperateCooperateTakeCooperateCooperateCooperate
Always takesTakes every round, whatever happens.TakeTakeTakeTakeTakeTake
Always cooperatesPlays fair every round, even when you take.CooperateCooperateCooperateCooperateCooperateCooperate
Never forgivesFair, until you take once. Then takes for ever.CooperateCooperateCooperateTakeTakeTake
RandomFlips a coin each round. Impossible to read.CooperateTakeCooperateCooperateTakeTake
Tit for TatFair first. Then does whatever you did last round.CooperateCooperateCooperateTakeCooperateCooperate

CCooperate: play fairTTake: grab more for yourself+3Points that round

And the winner was the simplest one.

Tit for Tat. Start fair, then copy what the other player did last round. Axelrod ran the competition again with 62 programs. Everyone knew Tit for Tat had won, and many tried to beat it. It won again.

  1. 01Tit for Tat
  2. 02
  3. 03
  4. 04
  5. 05
  6. 06
  7. 07
  8. 08
  9. 09
  10. 10
  11. 11
  12. 12
  13. 13
  14. 14
  15. 15
The first tournament, 15 programs in finishing order. Yellow: never took first Black: took first. The top eight never took first.
Watch the full storyVeritasium: Once you understand game theory, you understand the world.27 minutes on YouTube, with Robert Axelrod. Opens in a new tab. TLEP is not connected to Veritasium.

04 / How Tit for Tat works

Start fair. Respond straight away. Then forgive.

You play Tit for Tat. They take once, in round 3. Press Play, or go round by round.

Tit for Tat against a player who takes once in round 3
Round1234567Total
YouTit for TatCooperate, +3Cooperate, +3Cooperate, +0Take, +5Cooperate, +3Cooperate, +3Cooperate, +320
ThemTakes onceCooperate, +3Cooperate, +3Take, +5Cooperate, +0Cooperate, +3Cooperate, +3Cooperate, +320

  1. Round 1. You start fair. So do they. 3 points each.
  2. Round 2. Still fair. 3 each.
  3. Round 3. They take. You played fair, so you get 0 and they get 5.
  4. Round 4. You respond straight away: you take back. Now they get 0. Taking from you did not pay.
  5. Round 5. They played fair last round, so you forgive. Back to fair, 3 each.
  6. Round 6. Fair again. 3 each.
  7. Round 7. You both finish on 20. Taking got them nothing extra, and you never let it slide.

CCooperate: play fairTTake: grab more for yourself+3Points that round

  1. Be nice.

    Start fair. Never be the first to take.

  2. Respond.

    If someone takes from you, respond straight away. Once, not harder.

  3. Forgive.

    When they go back to playing fair, so do you. No grudges.

  4. Be clear.

    Be easy to read, so people know where they stand.

Responding does not mean violence.

It means not letting it pass: straight away, once, and in proportion. One take for one take, never more. Then, the moment they play fair again, you forgive. No strategy wins in every world, but across forty years of research the lesson has held: be nice, forgive, and don’t be a pushover.

Axelrod (1980a, 1980b); Axelrod and Hamilton (1981), Science; Rapoport, Seale and Colman (2015). The wording of the rules is TLEP’s plain-English reading.

05 / Why responding matters

Never responding is how you keep losing.

Here are two people up against someone who always takes. The first never responds. The second plays Tit for Tat.

Never responds

Always cooperating against always taking
Round123456Total
YouAlways cooperateCooperate, +0Cooperate, +0Cooperate, +0Cooperate, +0Cooperate, +0Cooperate, +00
ThemAlways takesTake, +5Take, +5Take, +5Take, +5Take, +5Take, +530

  1. Round 1. You play fair. They take. You get 0, they get 5.
  2. Round 2. You play fair again. They take again. 0 for you.
  3. Round 3. Same again.
  4. Round 4. Same again.
  5. Round 5. Same again.
  6. Round 6. You never responded, so they never stopped. You: 0. Them: 30.

Tit for Tat

Tit for Tat against always taking
Round123456Total
YouTit for TatCooperate, +0Take, +1Take, +1Take, +1Take, +1Take, +15
ThemAlways takesTake, +5Take, +1Take, +1Take, +1Take, +1Take, +110

  1. Round 1. You start fair. They take. 0 for you.
  2. Round 2. You respond straight away: you take too. 1 each.
  3. Round 3. They keep taking, so you keep protecting yourself. 1 each.
  4. Round 4. 1 each.
  5. Round 5. 1 each.
  6. Round 6. You lost once, then stopped being a target. You: 5. Them: 10, not 30.

CCooperate: play fairTTake: grab more for yourself+3Points that round

We teach the opposite.

What we tell people

Turn the other cheek. Be the bigger person. Sticks and stones. Keep your head down.

What the experiment found

Start fair. Respond straight away. Forgive. Be clear.

TLEP’s view: when people are told to absorb harm all their lives, without the skills or the support to respond, they keep losing, and the loss builds.

3 in 4

suicides in England and Wales are men. The male rate is three times the female rate.

ONS, 2023 and 2025 registrations

25%+

higher suicidal thinking among bullied teenagers who did nothing, compared with those who got help from an adult.

Benatov and others, Journal of School Violence, 2020, 11 countries. Association.

Research links men’s suicide to pressure to cope alone and not show what they feel. Being heard, and being able to respond to harm without violence, matters. These are links, not proof of cause.

06 / Mistakes

Mistakes start feuds. Forgiveness ends them.

In real life, people get things wrong. A look taken the wrong way. A message read wrong. A rumour. In the game that is a misread: someone played fair, but the other person saw it as taking. It is marked ! below.

A and B both play Tit for Tat. Watch what two misreads do.

Two Tit for Tat players with two misreads
Round123456789101112Total
ATit for TatCooperate, +3Cooperate, +3Cooperate, +3Cooperate, misread by the other player, +3Cooperate, +0Take, +5Cooperate, +0Take, +5Take, +1Take, +1Take, +1Take, +126
BTit for TatCooperate, +3Cooperate, +3Cooperate, +3Cooperate, +3Take, +5Cooperate, +0Take, +5Cooperate, misread by the other player, +0Take, +1Take, +1Take, +1Take, +126

  1. Round 1. Both play fair. 3 points each.
  2. Round 2. Both fair. 3 each.
  3. Round 3. Both fair. 3 each.
  4. Round 4. Both play fair, but B misreads A’s move as taking (the !).
  5. Round 5. B thinks A took, so B responds: B takes. A gets 0.
  6. Round 6. A sees B take, so A responds: A takes. Now B gets 0.
  7. Round 7. B responds to that. A gets 0 again. The echo has started: they take turns hitting back.
  8. Round 8. A responds again. B plays fair this round, but A misreads it as taking (the second !).
  9. Round 9. Now both take at the same time.
  10. Round 10. Both take. 1 point each.
  11. Round 11. Neither will play fair first. 1 each.
  12. Round 12. Stuck. Both finish on 26, when they could have had 36. That is a feud.

Now the same, but A lets it go.

A forgiving player and a Tit for Tat player with the same two misreads
Round123456789101112Total
AForgiving Tit for TatCooperate, +3Cooperate, +3Cooperate, +3Cooperate, misread by the other player, +3Cooperate, +0Cooperate, +3Cooperate, +3Cooperate, +3Cooperate, +3Cooperate, +3Cooperate, +3Cooperate, +333
BTit for TatCooperate, +3Cooperate, +3Cooperate, +3Cooperate, +3Take, +5Cooperate, +3Cooperate, +3Cooperate, misread by the other player, +3Cooperate, +3Cooperate, +3Cooperate, +3Cooperate, +338

  1. Round 1. Both play fair. 3 points each.
  2. Round 2. Both fair. 3 each.
  3. Round 3. Both fair. 3 each.
  4. Round 4. B misreads A’s fair move as taking (the !).
  5. Round 5. B responds by taking. A gets 0.
  6. Round 6. A lets it go: A plays fair instead of hitting back. B copies A’s fair move.
  7. Round 7. Back to fair. 3 each.
  8. Round 8. Fair, but A misreads B’s move as taking (the second !).
  9. Round 9. A lets it go again. Still fair.
  10. Round 10. Fair. 3 each.
  11. Round 11. Fair. 3 each.
  12. Round 12. A gave up a few points by forgiving, but both did far better than the feud: 33 and 38, instead of 26 each.

CCooperate: play fairTTake: grab more for yourself+3Points that round

When mistakes happen, a Tit for Tat that forgives now and then does even better than a strict one. Respond, yes. But leave a way back.

Wu and Axelrod (1995); Nowak and Sigmund (1992).

07 / Your turn

Now you try.

Ten rounds. Your choices go in the top row, theirs in the row underneath, with the points each round and your totals at the end. Try playing fair, try taking, and see how each strategy reacts.

1. Pick who you play against

2. Each round, choose. They choose at the same time.

Your game, round by round
Round12345678910Total
Youyour choices0
ThemCopycat0

CCooperate: play fairTTake: grab more for yourself+3Points that round

The game needs JavaScript. Everything it teaches is on this page.

08 / Where do you see this?

Where do you see this game?

This game is not really about points. It is about how we treat each other. So ask the same four questions of the places that shape a young person’s life. Do they start fair? When something goes wrong, do they respond clearly and in proportion, or not at all, or far too hard? Do they forgive and leave a way back? Are their rules clear?

  1. Schools

    When a young person gets it wrong, do we respond in proportion, or remove them?

    85% of children who offended had been suspended from school before their first offence. Pupils with special educational needs are 4 times as likely to be suspended.

    Ministry of Justice, May 2026; DfE, July 2026. Association.

  2. The justice system

    Do we respond once and fairly, and leave a way back?

    68% of adults leaving prison sentences under 12 months reoffend. The further a child goes into the formal system, the less likely they are to stop.

    Ministry of Justice, July 2026; McAra and McVie, 2007.

  3. Families under pressure

    When things went wrong at home, who responded, and who forgave?

    Adults with four or more harmful childhood experiences had 20 times the odds of going to prison.

    Public Health Wales, 2016. Association.

  4. Services and organisations

    Who took first: the young person, or the system around them?

    When London’s youth clubs closed, local 10 to 17 year olds became 14% more likely to offend.

    Institute for Fiscal Studies, 2024.

  5. Government

    Same game, same score?

    A £448 million programme for troubled families found no significant or systematic impact in its first phase. The NEET rate has barely crept below 10% in 25 years.

    DCLG, 2016; DWP Milburn review, 2026.

TLEP’s question

If almost everyone around you takes first, what would you learn to do? Maybe that is why the same problems keep coming back. Not because people are bad, but because the game they are in rewards taking.

09 / The town

Put good players together, and they spread.

Here is a town. Every square is a person. Most of the town always takes. A few people in the middle play Tit for Tat. Each round, everyone plays their eight neighbours, then copies whoever around them did best. Go round by round and watch.

Yellow: plays Tit for Tat (fair first, responds, forgives)Black: always takes

The town needs JavaScript to play. What happens: each round the yellow group grows by one ring of neighbours, because fair players who mostly meet each other score more, and the takers next to them copy them, until the whole town plays fair. If neighbours only ever play one game, the group cannot grow.

This is what Axelrod found: a small group who play fair and mostly meet each other can grow, even in a town of takers. In the long run, the players shape the place. That is why this starts from the ground up, close to home.

9%

lower murder rates for every 10 more local community organisations per 100,000 people, in a study of 264 US cities.

Sharkey and others, American Sociological Review, 2017

1 in 5

fewer custodial sentences at 16 for children who grew up within 2.5 km of a Sure Start centre.

Institute for Fiscal Studies, 2024

10 / The numbers that have not moved

Same game.
Same score.

981,000

young people aged 16 to 24 are not in education, employment or training: 13%. “The NEET rate has barely crept below 10% in 25 years.”

ONS, August 2026; DWP Milburn review, May 2026

50%

more likely to be NEET if you grew up disadvantaged, even with the same qualifications.

Impetus and NIESR, 2019

68%

of adults leaving prison sentences under 12 months reoffend. Overall reoffending has stayed between 23% and 32% since 2013.

Ministry of Justice, July 2026

42%

of prisoners were permanently excluded from school. Over half of permanently excluded children are later cautioned or sentenced.

Ministry of Justice, 2012 and May 2026. Association.

79%

of teenagers who carried a weapon had been victims of violence themselves that year.

Youth Endowment Fund, 2025

Half

of people entering prison may be neurodivergent, the inspectorates estimate, with patchy data and low staff awareness.

Criminal Justice Joint Inspection, 2021

What did move

Some numbers did change when the approach changed. Children cautioned or sentenced fell 65% in ten years after a national shift to diversion: responding to children without pulling them deeper into the formal system. Homicide in England and Wales is at its lowest rate since 1977. It was a national policy, which shows the rules can work at any level. TLEP reads it as the “forgive” rule in practice.

Youth Justice Statistics, January 2026; ONS, February 2026.

11 / So we are building the loop

The four rules,
built into a business.

  1. Nice.

    We offer first: paid work now, then room and kit.

  2. Respond.

    Clear boundaries, fair prices, honest feedback.

  3. Forgive.

    Second chances are the point.

  4. Clear.

    Every pound in the Open Book.

Our outcomes are not temporary jobs to tick boxes. They are skills, opportunity and experience towards a career people actually want.

See the loop Lend a hand
Watch the full storyVeritasium: Once you understand game theory, you understand the world.27 minutes on YouTube, with Robert Axelrod. Opens in a new tab. TLEP is not connected to Veritasium.

Every claim, sourced.

Figures checked on 18 October 2026. Links between things are marked as links, not proof that one causes the other. The game shows patterns in points, not people’s lives: it is a way of seeing, not a measured model of any community.

  1. Axelrod R (1980). Effective choice in the prisoner’s dilemma. Journal of Conflict Resolution 24(1): 3-25. Link 
  2. Axelrod R (1980). More effective choice in the prisoner’s dilemma. Journal of Conflict Resolution 24(3): 379-403. Link 
  3. Axelrod R, Hamilton WD (1981). The evolution of cooperation. Science 211: 1390-1396. Link 
  4. Nowak MA, Sigmund K (1992). Tit for tat in heterogeneous populations. Nature 355: 250-253. Link 
  5. Wu J, Axelrod R (1995). How to cope with noise in the iterated prisoner’s dilemma. Journal of Conflict Resolution 39(1): 183-189.
  6. Rapoport A, Seale DA, Colman AM (2015). Is tit-for-tat the answer? PLoS ONE 10(7): e0134128. Link 
  7. Veritasium (23 December 2023). Once you understand game theory, you understand the world. YouTube. Link 
  8. Office for National Statistics (27 August 2026). Young people not in education, employment or training (NEET), UK: August 2026. Link 
  9. Department for Work and Pensions (28 May 2026). Young people and work: interim report (independent review led by Alan Milburn). Link 
  10. Impetus and NIESR (April 2019). Establishing the Employment Gap. Link 
  11. Ministry of Justice (30 July 2026). Proven reoffending statistics: July to September 2024. Link 
  12. Ministry of Justice (March 2012). Prisoners’ childhood and family backgrounds. Link 
  13. Ministry of Justice (14 May 2026). Patterns of education, social care and youth offending. Link 
  14. Department for Education (9 July 2026). Suspensions and permanent exclusions in England: 2024/25. Link 
  15. Criminal Justice Joint Inspection (15 July 2021). Neurodiversity in the criminal justice system: a review of evidence. Link 
  16. Bellis MA and others (2016). Adverse Childhood Experiences and their impact on health-harming behaviours in the Welsh adult population. Public Health Wales. Link 
  17. Youth Endowment Fund (November 2025). Children, violence and vulnerability 2025. Link 
  18. Villa C (November 2024). The effects of youth clubs on education and crime. Institute for Fiscal Studies Working Paper 24/51. Link 
  19. Carneiro P and others (October 2024). The effect of Sure Start on youth misbehaviour, crime and contacts with children’s social care. Institute for Fiscal Studies. Link 
  20. Sharkey P, Torrats-Espinosa G, Takyar D (2017). Community and the crime decline. American Sociological Review. Link 
  21. Department for Communities and Local Government (October 2016). National Evaluation of the Troubled Families Programme: Final Synthesis Report. Link 
  22. Office for National Statistics (29 August 2024). Suicides in England and Wales: 2023 registrations. Link 
  23. Office for National Statistics (13 October 2026). Suicides in England and Wales: 1981 to 2025. Link 
  24. Benatov J and others (2020). Doing nothing is sometimes worse. Journal of School Violence 19(4): 456-469. Link 
  25. Youth Justice Board and Ministry of Justice (29 January 2026). Youth Justice Statistics 2024 to 2025. Link 
  26. Office for National Statistics (5 February 2026). Homicide in England and Wales: year ending March 2025. Link 
  27. McAra L, McVie S (2007). Youth justice? The impact of system contact on patterns of desistance from offending. European Journal of Criminology 4(3): 315-345. Link 

We use what you send to reply, and email you a copy. About your information.