Existence and Impossibility
To prove a process can never reach a state, find a quantity it preserves (parity, a sum mod , a coloring count) that differs between start and goal. To prove it terminates, find a monovariant — an integer quantity that strictly decreases. The 1..2026 difference game: the parity of the sum is invariant and odd, so the last number is odd — at least .
Look at the largest, smallest, or closest object — extremeness forces structure. Pigeonhole with objects in boxes forces together; the generalized and infinite versions and "pigeonhole on residues" cover a huge fraction of olympiad existence proofs.
Count incidences two ways to derive an identity or inequality (the handshake lemma is the prototype). Graph theory: degree sums, trees have edges, bipartite iff no odd cycle, Mantel/Turán bound edges without a triangle (; for ). Model the problem as a graph the moment "pairs" or "relations" appear.
Double Counting in Full
Let be a set of pairs (element, set), (vertex, edge), (point, line), and so on. Counting by its first coordinates and by its second gives the same number: . When one side can only be bounded, the equation becomes an inequality, and convexity finishes: since is convex, where is the average of the .
Write the incidences as a matrix with a row for each and a column for each . The total number of s is the sum of the row sums and also the sum of the column sums; that is all double counting is. Its power is in choosing : count the objects the problem cares about (triangles, common neighbours, pairs of committee members) through something you can control on the other side. The convexity step is Jensen for , whose second derivative is ; it turns a sum of into a statement about the average degree alone.
Compute .
Count pairs with . Grouped by , the count is the requested sum. Grouped by : each element lies in exactly subsets (the other elements are free). So the sum is ; for it is . A check from the other side: sizes occur times, and .
How to find it. A sum over sets of a size is a count of (element, set) memberships. Swap the order of summation.
In a tournament (every pair of players plays once, no draws) player has wins. Prove that the number of cyclic triangles ( beats beats beats ) is . For , what is the largest possible number?
Step 1 — a triangle that is not cyclic is transitive, and a transitive triangle has exactly one player who beat both others. So count pairs (player, two players she beat): grouped by player, there are ; grouped by triangle, every transitive triangle contributes exactly and every cyclic one . Hence the number of transitive triangles is and the rest, , are cyclic.
Step 2 — the sum to (each game produces one win), so their average is . By convexity . For the average is , so and there are at most cyclic triangles.
Step 3 — the bound is attained: seat the players around a circle and let each beat the next three clockwise. Everyone has wins, so exactly triangles are cyclic.
How to find it. Cyclic triangles are hard to count; transitive ones have a distinguished vertex. Count the complement through the vertex that makes it special.
A graph on vertices contains no -cycle. Prove that it has at most edges. (Reiman, 1958.)
Step 1 — count “cherries”: triples with distinct neighbours of . Grouped by , there are .
Step 2 — grouped by the pair : two distinct vertices both adjacent to and would give the -cycle . So each pair has at most one common neighbour, and .
Step 3 — with edges the average degree is , and convexity gives , i.e. . Multiply by : . The larger root of this quadratic in is , so is at most that. For the bound is , so at most edges.
How to find it. A forbidden subgraph with two paths of length two between the same ends: count paths of length two. The forbidden structure is exactly a statement about the other side of the count.
In Example 2 it is tempting to count transitive triangles by both their top and bottom players: tops plus bottoms. For the five-player tournament in which each player beats the next two around a circle, that gives “transitive triangles” — all of them — predicting no cyclic triangles. Yet beats , beats and beats : that triangle is cyclic, and in fact of the triangles are. The error: every transitive triangle has one top AND one bottom, so the second sum counts each of them again. Decide before counting what each object contributes, and make it exactly one.
Fifty people form committees of members each. Prove that someone serves on at least committees.
Show the answerHide the answer
Count (person, committee) memberships: by committees there are ; by people it is the sum of the numbers of committees per person. Their average is , and some number is at least the average. (If everyone served on at most , the total would be at most .)
Graphs as Models: Ramsey Arguments
If the edges of are colored red and blue, some triangle has all three edges the same color; can be colored with none. In symbols, . With three colors, every coloring of has a monochromatic triangle.
Pick a vertex of . Of its edges, at least share a color — pigeonhole, — say red edges to . If any edge among is red, it closes a red triangle with ; if none is, is a blue triangle. For , color the five sides of a pentagon red and the five diagonals blue: each color class is a -cycle, and a -cycle has no triangle. The same step with three colors: a vertex of has edges, so share a color; either that color appears among the neighbours (a triangle) or they span a in two colors.
Show that in any two-coloring of with no monochromatic triangle, every vertex has exactly two red and two blue edges, and each color class is a -cycle.
Step 1 — if a vertex had three edges of one color, say red to , the argument of the box gives a red triangle through or a blue triangle . So each vertex has at most two edges of each color, and with four edges, exactly two of each.
Step 2 — the red graph is then -regular on vertices: a union of cycles covering all vertices. A triangle is forbidden, and a -vertex -regular graph without a -cycle cannot split as (no -cycles exist), so it is a single -cycle. The blue graph is its complement, also a -cycle.
How to find it. Run the proof of backwards: every inequality in it must be tight, and tightness pins the structure.
Prove that every red/blue coloring of the edges of has at least two monochromatic triangles. (Goodman.)
Step 1 — call a pair of edges at a common vertex a bichromatic angle if they have different colors. A vertex with red and blue edges carries bichromatic angles, so there are at most in all.
Step 2 — count the same angles by triangles. A monochromatic triangle contains none; a triangle with two edges of one color and one of the other contains exactly (at the two ends of the odd edge). So twice the number of non-monochromatic triangles is at most : at most of them.
Step 3 — has triangles, so at least are monochromatic. The bound is sharp: color red and the two remaining triangles blue.
How to find it. Monochromatic triangles are what you want; bichromatic angles are what you can count at a vertex. Relate the two by counting angles triangle by triangle.
Seventeen scientists correspond pairwise, each pair about exactly one of three topics. Prove that some three of them correspond pairwise about the same topic. (IMO 1964/4.)
Step 1 — model: vertices are scientists, edges colored by topic, a -coloring of . Fix a scientist . Her edges carry three colors, so some color, say topic , is used at least times, to .
Step 2 — if two of the correspond about topic , they form a topic- triangle with .
Step 3 — otherwise all edges among use only topics and : a two-coloring of , which by the box contains a monochromatic triangle. (Sixteen scientists are not enough: a -coloring of without a monochromatic triangle exists — Greenwood and Gleason, 1955.)
How to find it. Reduce the number of colors by one: a large monochromatic neighbourhood either closes a triangle in that color or is a smaller Ramsey problem in the remaining colors.
“Among people there are mutual acquaintances or mutual strangers — proof: a person knows or does not know each of the other , so two of those relations are the same, and ...” The pigeonhole step gives only edges of one color at a vertex, and two edges never force a triangle. The statement itself is false: the pentagon-and-pentagram coloring of has no monochromatic triangle. Pigeonhole must produce three same-colored edges, which is why the threshold is .
Show that every red/blue coloring of has at least monochromatic triangles.
Show the answerHide the answer
A vertex with red and blue edges has bichromatic angles, so there are at most . Each non-monochromatic triangle contains exactly , so there are at most of them. Out of triangles, at least are monochromatic.