An Axiomatic Theory of Tie-Breaking: Impossibility, Characterization, and Decomposition

Frank M. V. Feys

arxiv: 2605.22846 · v1 · pith:JERHC22Znew · submitted 2026-05-15 · 💻 cs.GT · econ.TH· math.CO

An Axiomatic Theory of Tie-Breaking: Impossibility, Characterization, and Decomposition

Frank M. V. Feys This is my paper

Pith reviewed 2026-05-25 00:31 UTC · model grok-4.3

classification 💻 cs.GT econ.THmath.CO

keywords tie-breakinganonymitysymmetryorbit partitionaxiomatic theorysocial choicegame theory

0 comments

The pith

No tie-breaking rule producing a strict linear order can be anonymous if the input space contains even one intrinsically symmetric situation.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper builds an abstract framework for tie-breaking consisting of a finite player set N, a weak order on N, and auxiliary information equipped with a Sym(N) action. It proves that anonymity is impossible for any rule that always outputs a strict linear order whenever the input space includes at least one intrinsically symmetric situation, a condition satisfied in essentially all realistic applications. When the output is permitted to be a partition of N rather than a total order, exactly one rule satisfies two natural axioms of anonymity and neutrality: the partition of N into orbits of the joint stabilizer of the input. Every reasonable strict tie-breaking rule factors uniquely through this canonical orbit partition followed by an arbitrary completion to a linear order, separating the symmetry-respecting component from the necessary arbitrary choices.

Core claim

Within a framework consisting of a finite set N of players, a weak order on N, and auxiliary information on which Sym(N) acts, no anonymous tie-breaking rule can output a strict linear order if the input space contains at least one intrinsically symmetric situation. When partitions are permitted, the unique rule satisfying two natural axioms is the orbit partition of the joint stabilizer of the input. Every reasonable strict tie-breaking rule decomposes uniquely as the canonical orbit partition followed by an arbitrary completion.

What carries the argument

the orbit partition of the joint stabilizer of the input (the partition of N into orbits under the subgroup fixing both the weak order and the auxiliary information)

If this is right

Every reasonable strict tie-breaking rule decomposes uniquely into the canonical orbit partition followed by an arbitrary completion.
Real tie-breaking systems remain honest with respect to symmetry until forced to introduce arbitrary choices.
The same formalism captures chess tournament tie-breakers, sports league regulations, voting tie-breakers, symmetric players in cooperative games, and network centrality rankings.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

Existing tie-breaking procedures in sports or elections could be audited by checking whether they first respect the orbit partition before completing it arbitrarily.
If a domain genuinely lacks any symmetric input situations, the impossibility result would not apply and anonymous strict rules might exist.
The decomposition suggests a general pattern for analyzing other symmetry-breaking tasks where honest constraints must be separated from unavoidable arbitrary decisions.

Load-bearing premise

Every realistic application contains at least one intrinsically symmetric input situation and auxiliary information is always equipped with a Sym(N) action.

What would settle it

An explicit anonymous rule that always outputs a strict linear order on an input space containing at least one intrinsically symmetric situation would falsify the impossibility theorem.

Figures

Figures reproduced from arXiv: 2605.22846 by Frank M. V. Feys.

**Figure 1.** Figure 1: The two-level partition structure on N = {a, b, c, d, e, f}. The weak order ⪰ partitions N into indifference classes E1 (top) and E2 (bottom). Within each class, the orbit partition Ω(h, ⪰) subdivides further into blocks: E1 contains B1 = {a, b} and B2 = {c, d}; E2 contains B3 = {e} and B4 = {f}. The order between classes is forced by ⪰; the order of blocks within each class is the data γ; the order of pla… view at source ↗

read the original abstract

We develop an abstract axiomatic theory of tie-breaking. A tie-breaking input consists of a finite set N of players, a weak order on N representing the standings to be refined, and an auxiliary information item drawn from a set on which the symmetric group Sym(N) acts. Within this minimal framework we prove three theorems. First, no tie-breaking rule producing a strict linear order can be anonymous, provided the input space contains even one intrinsically symmetric situation, a condition met in essentially every realistic application. Second, when we allow the rule to output a partition of N (rather than a strict ranking), there is a unique rule satisfying two natural axioms: it is the partition of N into orbits of the joint stabilizer of the input. Third, every reasonable strict tie-breaking rule decomposes uniquely as the canonical orbit partition followed by an arbitrary completion. The decomposition makes precise the informal observation that real tie-breaking systems are honest until forced to be arbitrary. The framework is broad enough to capture chess tournament tie-breakers, sports league regulations, voting tie-breakers, tie-breaking among symmetric players in cooperative games, and ranking by network centrality measures, all within a single uniform formalism.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

The paper builds a Sym(N) group-action framework for tie-breaking and claims three theorems on anonymity impossibility, unique orbit partitions, and decomposition, but everything turns on the input space always containing symmetric cases.

read the letter

The core contribution is a minimal framework where tie-breaking inputs are a weak order plus auxiliary data with a Sym(N) action, then three results: no anonymous rule can output a strict linear order if there's even one symmetric input; the unique rule outputting a partition is the orbit partition of the joint stabilizer; and any reasonable strict rule factors uniquely into that orbit partition plus some completion. This setup is meant to cover chess tie-breakers, league rules, voting, cooperative games, and centrality measures in one language. The decomposition part does capture the informal point that real systems stay canonical until they are forced to pick arbitrarily. That is the part that feels most useful if the axioms turn out to be the right ones. The main soft spot is the precondition that every realistic domain has at least one intrinsically symmetric input with non-trivial joint stabilizer. The abstract asserts this holds in essentially every application but gives no exhaustive check against possible counter-domains, so both the impossibility and the uniqueness claims are conditional on that modeling choice. Without the full proofs or the exact axiom list it is also impossible to see how natural the two axioms for the partition rule really are or whether the notion of reasonable rules is tight. This is for readers in social choice or algorithmic game theory who already work with group actions or orbit decompositions. A serious referee could check whether the proofs close the gaps and whether the framework actually applies as broadly as claimed. I would send it to review rather than desk reject.

Referee Report

1 major / 1 minor

Summary. The manuscript develops an abstract axiomatic theory of tie-breaking. Inputs consist of a finite set N, a weak order on N, and auxiliary information equipped with a Sym(N) action. It proves three theorems: no anonymous rule outputting a strict linear order exists when the input space contains at least one intrinsically symmetric situation (a condition asserted to hold in essentially every realistic application); when outputs are allowed to be partitions, the unique rule satisfying two natural axioms is the partition into orbits of the joint stabilizer; and every reasonable strict tie-breaking rule decomposes uniquely into this canonical orbit partition followed by an arbitrary completion. The framework is presented as unifying chess tournaments, sports regulations, voting, cooperative games, and network centrality rankings.

Significance. If the framework and its modeling assumptions are accepted, the results supply a uniform group-theoretic characterization that isolates the determined component of tie-breaking from the necessarily arbitrary component. The decomposition theorem formalizes an informal observation about real systems and could guide the design of tie-breaking procedures that remain as canonical as possible. The approach is parameter-free once the action is fixed and yields falsifiable predictions about which rules can be anonymous.

major comments (1)

[Abstract, framework paragraph] Abstract (framework paragraph) and applications section: the impossibility, uniqueness, and decomposition theorems are all conditional on the input space containing at least one point whose joint stabilizer is non-trivial and on auxiliary data carrying a natural Sym(N) action. The manuscript asserts that this precondition holds 'in essentially every realistic application,' yet supplies no exhaustive argument that no counter-domain exists (e.g., domains in which auxiliaries lack a Sym(N) action or every input is asymmetric). This modeling choice is load-bearing for the scope of all three claims.

minor comments (1)

[Abstract] Abstract: the two natural axioms used for the uniqueness result are not stated explicitly; including their precise formulations would allow readers to assess the characterization immediately.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the careful review and for identifying the load-bearing modeling assumption in our framework. We address the major comment below.

read point-by-point responses

Referee: [Abstract, framework paragraph] Abstract (framework paragraph) and applications section: the impossibility, uniqueness, and decomposition theorems are all conditional on the input space containing at least one point whose joint stabilizer is non-trivial and on auxiliary data carrying a natural Sym(N) action. The manuscript asserts that this precondition holds 'in essentially every realistic application,' yet supplies no exhaustive argument that no counter-domain exists (e.g., domains in which auxiliaries lack a Sym(N) action or every input is asymmetric). This modeling choice is load-bearing for the scope of all three claims.

Authors: We agree that the three theorems are conditional on the stated precondition and that the manuscript does not supply an exhaustive argument ruling out all conceivable counter-domains. The framework is deliberately minimal and is intended for settings in which a natural Sym(N) action on auxiliary data exists (as occurs in the listed applications: chess tie-breakers via head-to-head results, sports regulations via tie-breaking criteria, voting with preference profiles, symmetric players in cooperative games, and centrality rankings). In each of these cases the auxiliary information is defined in a manner invariant under player relabeling, so the action is canonical. We do not claim the precondition is mathematically unavoidable in every abstract domain; rather, it is the modeling choice that makes the results applicable to the motivating examples. We will revise the abstract and applications section to state the precondition more explicitly as a modeling assumption required for the theorems and to avoid the phrasing 'essentially every realistic application' without qualification. revision: yes

Circularity Check

0 steps flagged

No circularity; theorems are standard consequences of group-action definitions within the stated framework.

full rationale

The paper defines a tie-breaking input as a weak order plus auxiliary data equipped with a Sym(N) action, then proves three results using the standard notions of joint stabilizer and orbits. The impossibility for strict linear orders follows directly from anonymity applied to a non-trivial stabilizer; the uniqueness for partitions is a characterization theorem identifying the orbit partition as the sole object satisfying the two axioms; the decomposition is the canonical splitting of any completion into the orbit partition plus an arbitrary linear extension. None of these steps reduces a derived quantity to a fitted parameter, renames a known empirical pattern, or relies on a load-bearing self-citation whose content is itself unverified. The framework assumptions are explicit and the derivations inside them are independent of the target claims.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The paper rests on the definition of the input space (finite N, weak order, auxiliary data with Sym(N) action) and on standard facts from group theory and order theory; no free parameters or new entities are introduced.

axioms (2)

domain assumption A tie-breaking input consists of a finite set N, a weak order on N, and auxiliary information on which Sym(N) acts.
This is the minimal framework stated in the abstract on which all three theorems depend.
domain assumption The input space contains at least one intrinsically symmetric situation.
Required for the anonymity impossibility; stated as met in essentially every realistic application.

pith-pipeline@v0.9.0 · 5741 in / 1422 out tokens · 55163 ms · 2026-05-25T00:31:52.688896+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

14 extracted references · 14 canonical work pages

[1]

Arrow,Social Choice and Individual Values, 2nd ed., Cowles Foundation Mono- graph 12, Yale University Press, New Haven, 1963

Kenneth J. Arrow,Social Choice and Individual Values, 2nd ed., Cowles Foundation Mono- graph 12, Yale University Press, New Haven, 1963

work page 1963
[2]

3, 1463–1483

Laurent Bartholdi, Wade Hann-Caruthers, Maya Josyula, Omer Tamuz, and Leeat Yariv, Equitable voting rules, Econometrica89(2021), no. 3, 1463–1483

work page 2021
[3]

2, 377–401

Daniela Bubboloni and Michele Gori,Anonymous and neutral majority rules, Social Choice and Welfare43(2014), no. 2, 377–401

work page 2014
[4]

Daniela Bubboloni and Michele Gori,Resolute refinements of social choice correspondences, Mathematical Social Sciences84(2016), 37–49

work page 2016
[5]

1, 411–457

Daniela Bubboloni and Michele Gori,Breaking ties in collective decision-making, Decisions in Economics and Finance44(2021), no. 1, 411–457

work page 2021
[6]

1–2, 17–36

László Csató,On the ranking of a Swiss-system chess team tournament, Annals of Operations Research254(2017), no. 1–2, 17–36

work page 2017
[7]

3, 665–681

Allan Gibbard,Manipulation of schemes that mix voting with chance, Econometrica45 (1977), no. 3, 665–681

work page 1977
[8]

Procaccia (eds.), Handbook of Computational Social Choice, Cambridge University Press, Cambridge, 2016

Felix Brandt, Vincent Conitzer, Ulle Endriss, Jérôme Lang, and Ariel D. Procaccia (eds.), Handbook of Computational Social Choice, Cambridge University Press, Cambridge, 2016. 30

work page 2016
[9]

Hervé Moulin,The Strategy of Social Choice, Advanced Textbooks in Economics 18, North- Holland, Amsterdam, 1983

work page 1983
[10]

Saari,Basic Geometry of Voting, Springer, Berlin, 1995

Donald G. Saari,Basic Geometry of Voting, Springer, Berlin, 1995

work page 1995
[11]

Julius Petersen,Die Theorie der regulären graphs, Acta Mathematica15(1898), 193–220

work page
[12]

Shapley,A value forn-person games, inContributions to the Theory of Games, Volume II(H

Lloyd S. Shapley,A value forn-person games, inContributions to the Theory of Games, Volume II(H. W. Kuhn and A. W. Tucker, eds.), Annals of Mathematics Studies 28, Princeton University Press, Princeton, 1953, pp. 307–317

work page 1953
[13]

1133–1162

Lirong Xia,Most equitable voting rules, in Proceedings of the 24th ACM Conference on Economics and Computation (EC ’23), 2023, pp. 1133–1162

work page 2023
[14]

Lirong Xia,Computing most equitable voting rules, in Proceedings of the 20th Conference on Web and Internet Economics (WINE ’24), 2024. 31

work page 2024

[1] [1]

Arrow,Social Choice and Individual Values, 2nd ed., Cowles Foundation Mono- graph 12, Yale University Press, New Haven, 1963

Kenneth J. Arrow,Social Choice and Individual Values, 2nd ed., Cowles Foundation Mono- graph 12, Yale University Press, New Haven, 1963

work page 1963

[2] [2]

3, 1463–1483

Laurent Bartholdi, Wade Hann-Caruthers, Maya Josyula, Omer Tamuz, and Leeat Yariv, Equitable voting rules, Econometrica89(2021), no. 3, 1463–1483

work page 2021

[3] [3]

2, 377–401

Daniela Bubboloni and Michele Gori,Anonymous and neutral majority rules, Social Choice and Welfare43(2014), no. 2, 377–401

work page 2014

[4] [4]

Daniela Bubboloni and Michele Gori,Resolute refinements of social choice correspondences, Mathematical Social Sciences84(2016), 37–49

work page 2016

[5] [5]

1, 411–457

Daniela Bubboloni and Michele Gori,Breaking ties in collective decision-making, Decisions in Economics and Finance44(2021), no. 1, 411–457

work page 2021

[6] [6]

1–2, 17–36

László Csató,On the ranking of a Swiss-system chess team tournament, Annals of Operations Research254(2017), no. 1–2, 17–36

work page 2017

[7] [7]

3, 665–681

Allan Gibbard,Manipulation of schemes that mix voting with chance, Econometrica45 (1977), no. 3, 665–681

work page 1977

[8] [8]

Procaccia (eds.), Handbook of Computational Social Choice, Cambridge University Press, Cambridge, 2016

Felix Brandt, Vincent Conitzer, Ulle Endriss, Jérôme Lang, and Ariel D. Procaccia (eds.), Handbook of Computational Social Choice, Cambridge University Press, Cambridge, 2016. 30

work page 2016

[9] [9]

Hervé Moulin,The Strategy of Social Choice, Advanced Textbooks in Economics 18, North- Holland, Amsterdam, 1983

work page 1983

[10] [10]

Saari,Basic Geometry of Voting, Springer, Berlin, 1995

Donald G. Saari,Basic Geometry of Voting, Springer, Berlin, 1995

work page 1995

[11] [11]

Julius Petersen,Die Theorie der regulären graphs, Acta Mathematica15(1898), 193–220

work page

[12] [12]

Shapley,A value forn-person games, inContributions to the Theory of Games, Volume II(H

Lloyd S. Shapley,A value forn-person games, inContributions to the Theory of Games, Volume II(H. W. Kuhn and A. W. Tucker, eds.), Annals of Mathematics Studies 28, Princeton University Press, Princeton, 1953, pp. 307–317

work page 1953

[13] [13]

1133–1162

Lirong Xia,Most equitable voting rules, in Proceedings of the 24th ACM Conference on Economics and Computation (EC ’23), 2023, pp. 1133–1162

work page 2023

[14] [14]

Lirong Xia,Computing most equitable voting rules, in Proceedings of the 20th Conference on Web and Internet Economics (WINE ’24), 2024. 31

work page 2024