Politics is usually argued rather than measured, which makes this group unusual. It holds the findings that survive regardless of who is in power: mathematical limits on collective choice, patterns in how voters behave, and a body of criminology precise enough to direct police patrols street by street.
Limits no constitution can escape
The most striking results are proofs. Arrow's Impossibility Theorem shows that no voting rule for three or more options can satisfy a short list of reasonable conditions at once. The Gibbard–Satterthwaite Theorem adds that every non-dictatorial rule can be gamed by tactical voting, and Sen's Liberal Paradox shows that even minimal individual rights conflict with unanimous preferences. The Median Voter Theorem and the Coase Theorem are the constructive results: under stated conditions, parties converge on the centre, and bargaining finds efficient outcomes whoever holds the rights.
How voters and publics behave
The voting findings describe an electorate that judges results more than programmes. Retrospective Voting and Economic Voting find incumbents rewarded or punished for recent conditions. The Rally 'Round the Flag Effect finds support jumping at the start of a crisis. Thermostatic Public Opinion finds publics pushing back against whichever direction policy has recently moved. Duverger's Law connects voting rules to party systems: plurality elections in single-member districts tend to produce two parties.
Where crime happens
The largest cluster is criminology, and its central finding is concentration. The Law of Crime Concentration finds a small share of street segments producing a large share of crime in city after city. Repeat and Near-Repeat Victimization find that a place or person once victimised is at raised risk soon after, and the Victim-Offender Overlap finds the two populations largely the same. The Hot-Spots Effect and the Diffusion of Crime-Control Benefits show that focusing on those places reduces crime without simply pushing it round the corner. Beccaria's Certainty-Severity Asymmetry, stated in 1764, still holds: the likelihood of being caught deters far more than the harshness of the sentence.
Where it applies
The criminology is directly usable, and many police forces now allocate patrols by it. The voting results inform electoral reform, since Arrow and Duverger between them explain why every voting system has trade-offs and why changing the rules changes the parties. The Iron Law of Prohibition and the Hydra Effect are standard considerations in drug policy: harder enforcement pushes suppliers towards more potent products, and removing a network's leaders can fragment it into more violent pieces. The Procedural Justice Effect shows that people comply more with authorities they see as fair, which matters to courts and police alike.
Where it is argued over
The voting findings carry an interpretive dispute. Achen and Bartels argued that voters punish incumbents even for droughts and shark attacks, which suggests retrospective judgement is less about performance than its defenders hoped. The Modernization Hypothesis is contested: Przeworski and colleagues found that wealth helps democracies survive more than it helps them emerge. Higgs's Ratchet and Director's Law are argued over along political lines as much as statistical ones.
Why it is worth knowing
This group sets the boundaries within which political argument takes place. It says that no voting system is perfect, that voters will judge results rather than intentions, and that crime is concentrated enough to be addressed with precision rather than with blanket severity. Those findings temper idealism and cynicism alike, which is useful in a field where both are plentiful.
The entries behind this essay: Evidence in Law & Politics. Every subject has its essays on the essays page.