← EisatoponAI← EisatoponAI
EisatoponAI · Strategy Map

Mathematical
Problem-Solving
Strategies

Twelve fundamental strategies used by olympiad mathematicians — from the Extremal Element to Mathematical Induction. Each comes with worked examples at three levels of difficulty.

12 Strategies·36 Problems·3 Levels each
01

Extremal Element

Start from the most constrained element

Finite SetsOptimizationContradiction

Instead of looking at all elements simultaneously, pick the most extreme one — the largest, smallest, first, or last. The extremal element has fewer possibilities than the rest, and that constraint gives you information.

Explore Strategy
02

Special & Boundary Cases

Test the extremes first

ParametersIntuitionTesting

Before attempting the general case, try the extremes. A special case occurs when a variable takes an extreme value: $n=1$, $x=0$, $k=n$, etc. These cases often reveal whether a claim is true and how to prove it.

Explore Strategy
03

Counterexamples

One example says YES — one counterexample says NO

DisproveFalse ClaimsSearch

A counterexample is the most direct method to disprove a claim. Instead of explaining why something is wrong, simply find one instance where it fails. One example is enough to refute a universal claim.

Explore Strategy
04

Backward Thinking

Start from the goal — work backwards

Goal-OrientedConstructionInequalities

Like solving a maze from the exit! Instead of starting from the data and trying to reach the conclusion, start from the conclusion and ask: 'What must be true for this to hold?' Often the goal has more structure than the hypotheses.

Explore Strategy
05

Symmetry

If it looks the same from everywhere, use it

WLOGSymmetric FunctionsGeometry

When a problem has symmetry, some variables 'play the same role'. This often simplifies it dramatically. If $x$, $y$, $z$ play the same role, the extremum must be symmetric. Three types: algebraic, geometric, cyclic.

Explore Strategy
06

Invariants

Find what remains constant when everything changes

ProcessesImpossibleParity

An invariant is a quantity that doesn't change while something else does. If the start and end states have different invariant values, the transformation is impossible. Invariants can be numerical, modular, or structural.

Explore Strategy
07

Pigeonhole Principle

More pigeons than holes → some hole has two pigeons

ExistenceCountingCombinatorics

If $n+1$ objects are placed in $n$ categories, at least one category contains $\geq 2$ objects. Simple to state, but its applications are astonishing. Generalized: if $kn+1$ objects go in $n$ categories, some has $\geq k+1$.

Explore Strategy
08

Double Counting

Count the same thing in two different ways

BijectionCombinatoricsGraph Theory

Count the same quantity from two perspectives, then equate the results. This yields a useful relation. Classic: counting handshakes by persons (sum of degrees) or by pairs (twice the number of handshakes).

Explore Strategy
09

Reformulation

Say the same thing in a different way

SubstitutionTranslationEquivalent Problems

Transform a problem into an equivalent one that is easier to solve. Change variables, translate to a different domain, or reinterpret the question. The key insight: if two problems are equivalent, solving either one solves both.

Explore Strategy
10

Auxiliary Construction

Create something new to solve the problem

GeometryExtra ElementsConnections

Add a new element to the problem — a line, point, function, or set — that wasn't in the original statement but helps connect the pieces and reach the solution.

Explore Strategy
11

Proof by Contradiction

Assume the opposite — then derive a contradiction

Indirect ProofNegationImpossibility

Assume the opposite of what you want to prove, then show this assumption leads to something impossible (a contradiction). If the negation is false, the original statement must be true. One of the most powerful and widely-used proof techniques.

Explore Strategy
12

Mathematical Induction

The ladder to infinity: one step at a time

All nRecursiveSequences

Prove something holds for infinitely many cases with just two steps. Base case: verify for $n_0$. Inductive step: show that if it holds for $n = k$, it holds for $n = k+1$. Then it holds for all $n \geq n_0$.

Explore Strategy