Feys' Axiomatic Tie-Breaking Theory Proves Impossibility & Unique Rule
No fair tie-breaker exists when players are symmetric – unless you relax the goal.
Get AI news that actually matters
One email a day. Zero fluff. Join 10,000+ professionals.
Frank Feys presents a rigorous mathematical framework for tie-breaking in a 31-page paper submitted to Social Choice and Welfare. The input consists of a finite set of players, a weak order (standings), and auxiliary information on which the symmetric group acts. Feys first proves an impossibility: no tie-breaking rule that produces a strict linear order can be anonymous if the input space contains even one intrinsically symmetric situation—a condition met in virtually all real-world applications.
Second, when the rule is allowed to output a partition instead of a strict ranking, a unique rule satisfies two natural axioms: it partitions players into orbits of the joint stabilizer of the input. Third, every reasonable strict tie-breaking rule uniquely decomposes into that canonical orbit partition followed by an arbitrary completion. This decomposition formalizes the intuition that real systems are honest until forced to be arbitrary. The theory applies across chess tournaments, sports league regulations, voting tie-breakers, cooperative games, and network centrality measures.
- No anonymous tie-breaking rule can produce a strict ranking in any realistic symmetric situation (impossibility theorem).
- The unique rule that outputs a partition uses joint stabilizer orbits of the input (characterization).
- Every strict tie-breaking rule decomposes into an honest partition step plus an arbitrary completion (decomposition).
Why It Matters
Provides a rigorous foundation for designing tie-breakers in tournaments, voting, and algorithmic rankings.