!
THE WRONG VICTORY
BLINDSPOT · INSIDE-OUT ECONOMICS
01 · THE CLASS VOTE

Rape is rising. How harsh should the law be?

Murder carries 40 years. Choose a rape sentence. These forecasts show only the first offence in 240 high-risk fictional cases.

A
MAXIMUM CRACKDOWN
40 years for rape · the harshest option
—simulated rapes
in 240 high-risk cases
B
TOUGH SENTENCE
25 years for rape
—simulated rapes
in 240 high-risk cases
C
SHORTER SENTENCE
10 years for rape
—simulated rapes
in 240 high-risk cases
AUDIENCE RESPONSE
Which policy makes the country safer?
Enter actual A, B and C votes.
—
—
—
Enter the class vote counts, or try demo votes.
Illustrative high-risk cases, not a general-population rape rate. Escalation outcomes are hidden.
02 · THE FIRST RESULT

Your law looks like a winner. Run the decisions.

240 deliberately high-risk fictional situations, not 240 people drawn from the public. First count rapes. Then reveal murders following rape.

READY · 240 FICTIONAL DECISIONS
No rapeRape
—Situations without rape
—Rape decisions
HIDDENFollowed by murder
CLASS LAW
—
—
RAPE DECISIONS · OF 240
—
See the initial-offence result first.
Fictional high-risk situations. Counts are not real crime rates or causal estimates.
03 · THE FIRST ESCALATION

You deterred the first crime. Did you protect the victim?

Change the sentence. Watch which of the same 240 high-risk decisions end in no rape, rape alone, or rape followed by murder.

CHOOSE THE RAPE SENTENCE
RAPE
40y
→
RAPE + MURDER
40y
EXTRA PENALTY
FOR MURDER+0y
The class law loads first.
Illustrative probabilities of conviction: 34% after rape, 20% after rape followed by murder. Not observed estimates.
THE SAME 240 FICTIONAL DECISIONS
NO RAPE
—No offence
RAPE ONLY
—No murder
RAPE + MURDER
—Victims killed
13.6y → 8yAssumed expected sentence cost: rape only → rape + murder, accounting for detection risk
See the difference when murder adds no extra sentence.
These are assumptions about behaviour under hypothetical sentencing rules. Actual outcomes cannot be inferred from them.
04 · SECOND EXPERIMENT · ONE STUDENT DECIDES

Does execution stop murder? Or move it?

Choose a law, then follow the same 240 fictional high-risk cases through three decisions: first murder → another murder → police officer killed.

ONE STUDENT · CHOOSE THE LAW
THE CHALLENGE
Will the harshest law stop the first murder—but remove the extra penalty for the next one?
All three laws face identical synthetic cases. Counts illustrate mechanisms, not observed homicide rates.
POLICY B · EXECUTE FOR ANY MURDERREADY
1 · FIRST MURDER Does the initial threat deter killing?—
2 · ANOTHER MURDER Of those who killed once, who kills again?HIDDEN
3 · POLICE OFFICER KILLED Does anyone kill an officer as well?HIDDEN
First reveal initial murders. Then see whether the same offenders cause further deaths.
05 · REVEAL

Reveal. What does the next murder cost?

Both experiments expose the same omission: the penalty for a first crime is not the penalty for doing more harm.

RAPE
→
RAPE + MURDER
Same 40-year ceiling?
FIRST MURDER
→
ANOTHER MURDER / POLICE DEATH
Execution already the maximum?
01
DETER THE FIRST CRIME
A tougher penalty can raise the expected cost of offending. Some first offences may be prevented.
02
NOW ASK: WHAT DOES THE NEXT CRIME ADD?
At the punishment ceiling, another murder may add zero extra statutory penalty. The marginal legal cost of escalation shrinks.
03
MARGINAL DETERRENCE
If the additional expected cost of further harm falls, an offender has less incentive to stop. Conviction risk and other costs still matter.
MAXIMUM PUNISHMENT ≠ MAXIMUM DETERRENCE

Marginal deterrence depends on the extra expected penalty for causing extra harm, not merely the total punishment.

The simulated choices show how escalation could occur, not that execution raises real-world murder rates. Zero additional statutory sentence is not necessarily zero marginal expected or personal cost.