📚 The Election Cycle: Condorcet Paradox and Binomial Voting Models | 选举周期：孔多塞悖论与二项投票模型
In democratic elections, voters express preferences among candidates. Usually, we expect a clear winner, but sometimes preferences can lead to a paradoxical situation: candidate A beats B, B beats C, yet C beats A. This is known as a voting cycle or the Condorcet paradox. Understanding such cycles requires mathematical reasoning from graph theory, while predicting election outcomes often uses probability models like the binomial distribution—core topics in A-Level Mathematics.
在民主选举中，选民表达对候选人的偏好。通常我们期待清晰的胜者，但有时偏好会导致矛盾：候选人A击败B，B击败C，而C又击败A。这就是投票循环或孔多塞悖论。理解这种循环需要图论中的数学推理，而预测选举结果则常用二项分布等概率模型——这些正是A-Level数学的核心内容。
1. Introduction to Voting Cycles | 投票循环简介
In many voting systems, each voter ranks candidates in order of preference. When pairwise comparisons are made, we may discover that there is no candidate who wins against every other candidate. Instead, a cyclical majority can emerge. This was first highlighted by the Marquis de Condorcet in the 18th century. The existence of such cycles challenges the notion that voting always yields a consistent and 'fair' outcome.
在许多投票制度中，每位选民按偏好顺序排列候选人。当成对比较时，我们可能发现没有一个候选人能击败所有其他候选人。相反，可能出现循环多数。这一问题最早由孔多塞侯爵在18世纪提出。这种循环的存在挑战了投票总能产生一致且“公平”结果这一观念。
The election cycle is not just a theoretical curiosity; it has practical implications for designing voting rules, from school council elections to parliamentary systems. Mathematically, it can be studied using directed graphs and preference matrices.
选举周期不仅是理论上的奇闻，它对从学生会选举到议会制度等投票规则的设计都有实际影响。在数学上，我们可以用有向图和偏好矩阵来研究它。
2. Preference Relations and Transitivity | 偏好关系与传递性
Preference between candidates can be modelled as a binary relation. If a voter prefers X over Y, we denote this as X ≻ Y. In an ideal rational world, preferences are transitive: if X ≻ Y and Y ≻ Z, then X ≻ Z. However, when aggregating individual preferences to form a societal preference, transitivity can break down.
候选人之间的偏好可以建模为一个二元关系。如果某选民偏好X胜过Y，我们记作X ≻ Y。在理想的理性世界中，偏好具有传递性：如果X ≻ Y且Y ≻ Z，则X ≻ Z。然而，当将个体偏好汇总为社会偏好时，传递性可能失效。
This failure is precisely what creates an election cycle. The Condorcet paradox provides a classic counterexample showing that majority rule can violate transitivity.
这种失效正是产生选举周期的原因。孔多塞悖论提供了一个经典反例，表明多数决可能违背传递性。
3. The Condorcet Paradox: An Example | 孔多塞悖论：实例
Consider three candidates: Alice, Bob, and Carol. Suppose 30% of voters rank Alice ≻ Bob ≻ Carol, 35% rank Bob ≻ Carol ≻ Alice, and 35% rank Carol ≻ Alice ≻ Bob. In pairwise comparisons, Alice defeats Bob (65% to 35%), Bob defeats Carol (65% to 35%), but Carol defeats Alice (70% to 30% orcet winner. This illustrates that simple majority rule can produce an intransitive social ordering.
考虑三位候选人：Alice、Bob和Carol。假设30%的选民排序Alice ≻ Bob ≻ Carol，35%排序Bob ≻ Carol ≻ Alice，35%排序Carol ≻ Alice ≻ Bob。在成对比较中，Alice击败Bob（65%对35%），Bob击败Carol（65%对35%），但Carol却击败Alice（70%对30%）。多数偏好形成了A → B → C → A的循环，意味着不存在孔多塞胜者。这表明简单的多数决可能产生非传递的社会排序。
This example uses only three candidates, but cycles can involve any number. The probability of such cycles occurring increases with the number of candidates and the diversity of preference orders.
该例仅涉及三位候选人，但循环可涉及任意数量。这种循环发生的概率随候选人数量和偏好顺序多样性的增加而增大。
4. Representing Preferences with Directed Graphs | 用有向图表示偏好
We can visualise pairwise outcomes using a directed graph (digraph). Each candidate is a vertex. A directed edge from X to Y indicates that a majority prefers X over Y. The Condorcet paradox corresponds to a directed cycle in this graph.
我们可以用有向图(digraph)来可视化成对比较的结果。每个候选人是一个顶点。从X到Y的有向边表示多数人偏好X胜过Y。孔多塞悖论对应于该有向图中的一个有向循环。
orming a 3-cycle. This is the simplest directed cycle possible. In decision mathematics, detecting cycles in a digraph is important for determining if a consistent ranking exists.
在我们的例子中，顶点A、B、C，有向边A→B、B→C、C→A，构成一个三元环。这是可能的最简单有向环。在决策数学中，检测有向图中是否存在环对于确定是否存在一致的排序至关重要。
5. Detecting Cycles: Graph Theory Basics | 检测循环：图论基础
From the A-Level decision maths perspective, we can use algorithms like depth-first search or check the adjacency matrix of the tournament (where each pair has exactly one directed edge). A tournament is transitive if and only if it contains no directed cycles of length 3 (or any length). However, the presence of a 3-cycle indicates intransitivity.
从A-Level决策数学的角度，我们可以使用类似于深度优先搜索的算法，或检查竞赛图（每对顶点之间有且仅有一条边）的邻接矩阵。一个竞赛图是传递的当且仅当它不包含任何长度的有向环（特别是三元环）。然而，存在3元环就是非传递的标志。
j, else 0. For our cycle, M = [[0,1,0],[0,0,1],[1,0,0]] if we order A,B,C. Noticing that M2 and M3 can reveal cycles: the trace of M3 gives the number of 3-cycles. This connects to core A-Level Further Maths topics like matrices.
考虑邻接矩阵M，若i击败j则M[i][j]=1，否则0。对我们的循环，若顺序为A,B,C，则M = [[0,1,0],[0,0,1],[1,0,0]]。注意到M²和M³能揭示环：M³的迹给出三元环的个数。这与A-Level进阶数学的核心主题矩阵相关联。
A simpler method: if we sort candidates by their number of 'wins' (out-degree), we may find a ranking, but cycles prevent a total order. The graph-theoretic approach shows why some voting systems must adopt tie-breaking rules or sequential elimination to arrive at a unique winner.
一种更简单的方法：如果我们按“获胜次数”（出度）对候选人排序，也许能得出一个排名，但循环阻碍了全序。图论方法说明了为何一些投票制度必须采用平局解消规则或逐轮淘汰来产生唯一胜者。
6. Breaking the Cycle: Voting Rules | 打破循环：投票规则
Various voting rules have been proposed to break or avoid cycles. For example, the Borda count assigns points based on ranking positions and sums them; this usually yields a transitive ordering but can violate the Condorcet criterion. Another rule is the instant-runoff voting (IRV), which eliminates the candidate with the fewest first-choice votes and redistributes ballots. Arrow's impossibility theorem states that no voting system can simultaneously satisfy a set of seemingly reasonable fairness criteria when there are three or more candidates.
人们提出多种投票规则来打破或避免循环。例如，波达计数法基于排名位置分配分数并加总；它通常产生传递排序，但可能违反孔多塞准则。另一种规则是即时决选投票(IRV)，淘汰首选票最少的候选人并重新分配选票。阿罗不可能定理指出，当有三位或更多候选人时，没有任何投票制度能同时满足一套看似合理的公平标准。
Understanding these rules mathematically helps evaluate their strengths and weaknesses. For instance, Borda count can be expressed as a weighted sum of pairwise wins. The mathematics of voting is a rich field intersecting with combinatorics and optimisation.
从数学上理解这些规则有助于评估其优劣。比如，波达计数可表示为成对获胜的加权和。投票数学是一个与组合数学和优化交叉的丰富领域。
7. Modelling Voter Behaviour: Binomial Probability | 选民行为建模：二项概率
While voting cycles deal with aggregated preferences, predicting the outcome of a single-winner election often uses probability models. Assume each voter independently chooses candidate X with probability p, and there are n voters. The number of votes for X follows a binomial distribution B(n, p). This is a standard topic in A-Level Mathematics: X ~ B(n, p).
虽然投票循环涉及汇总偏好，但预测单一胜者选举的结果常用概率模型。假设每位选民独立地以概率p投票给候选人X，有n位选民。那么投给X的票数服从二项分布 B(n, p)。这是A-Level数学的标准课题：X ~ B(n, p)。
We can calculate the probability that X wins if the election is simply 'who gets more than half the votes'. For a large n, the normal approximation can also be applied, using continuity correction—a technique required in exams.
若选举仅以“谁获半数以上选票”定胜负，我们可计算X获胜的概率。对于大n，也可使用正态近似，并运用连续性校正——这是考试要求的技术。
Published by TutorHao | A-Level Mathematics Revision Series | aleveler.com
广告 / Sponsored现货 Cambridge International AS & ALevel IT Coursebook剑桥国际IGC…计算机 备考教辅 · 正版现货 · 与本文内容相关查看详情 / View on Taobao参考价 ¥467.0
更多咨询请联系16621398022（同微信）
或截图后用「扫一扫 → 相册」选择本图