15 Math Project Ideas for High School Students in 2026
Most math project ideas fail at the same point. A student picks a title, builds a neat file about it, and never actually does any mathematics. The history of pi, the golden ratio in nature, Pythagoras in daily life: these are presentations about maths rather than projects that use it. Teachers marking internal assessments and universities reading applications look for the same signal, did the student compute something they did not already know the answer to?
The fifteen ideas below are questions with a computable answer. Each names the mathematics it needs and where the data comes from, so you can judge quickly whether it is realistic for you.
First, decide which kind of math project you are building
The same topic can become three very different pieces of work. Choose the tier first, because it decides how much evidence you owe the reader.
Tier | Time | What it must contain |
|---|---|---|
School file or exhibition project | 2–3 weeks | Correct mathematics, clear presentation, one original worked example |
Investigation | 6–10 weeks | Data you collected or sourced, a method applied, limitations stated |
Research paper | 4–6 months | A question unanswered in this form, a defensible method, referenced literature |
Students usually aim at tier one, then feel the project is thin. Moving up rarely means a harder topic, just the same topic with real data attached.
15 math project ideas for high school students
Probability and statistics you can collect yourself
These need a notebook, a spreadsheet, and patience. The mathematics sits inside the Grade 9–12 syllabus;
# | Question | Mathematics used | Data source |
|---|---|---|---|
1 | Do first digits of town population figures in your state follow Benford's Law? | Log distributions, chi-square goodness of fit | Census tables |
2 | Are arrivals at the school canteen a Poisson process? | Poisson distribution, arrival rate | Tally sheet, ten lunch breaks |
3 | Are the dice in a board-game set actually fair? | Expected vs observed frequency, chi-square | 600 recorded rolls |
4 | Does the birthday problem's prediction hold across your school's sections? | Combinatorics, complementary probability | Class registers, shared dates |
5 | Which sampling method estimates a textbook's word count most accurately? | Random, systematic, stratified sampling; confidence intervals | One full manual count |
Idea 5 is underrated: you end up with a true value, three estimates, and a finding about which method wins.
Geometry, trigonometry and measurement
# | Question | Mathematics used | Data source |
|---|---|---|---|
6 | How far do three height-measuring methods disagree, and why? | Trigonometry, percentage error, error propagation | Clinometer, shadow ratio, phone app |
7 | Which tilings are geometrically possible, and which appear locally? | Interior angle sums, symmetry, proof | Photographs of floors, facades |
8 | Do building facades really cluster near the golden ratio? | Ratio data, histograms, baseline comparison | Thirty facades, scaled photographs |
9 | Do commercial packages use the minimum surface area for their volume? | Optimisation, surface area and volume | Twenty packages off a shelf |
Idea 8 is worth flagging. Almost every list of mathematics project ideas asserts the golden ratio is everywhere. Testing that claim, and reporting honestly if your measurements do not support it, beats repeating it.
Modelling real systems
# | Question | Mathematics used | Data source |
|---|---|---|---|
10 | Which curve fits your city's monthly temperature record best? | Linear, quadratic, sinusoidal regression; R² | Meteorological archives |
11 | Does a recurring deposit actually grow after inflation? | Geometric sequences, real vs nominal returns | Published deposit, inflation rates |
12 | How fast would a rumour move through your school? | Discrete SIR model in a spreadsheet | Contact-rate survey, with consent |
13 | What green-light duration clears one intersection fastest? | Rate equations, queue length, simulation | Timed vehicle counts, one week |
Idea 12 is epidemic modelling made accessible: weekly spreadsheet steps rather than differential equations, and the same characteristic curve.
Discrete mathematics, graphs and cryptography
# | Question | Mathematics used | Data source |
|---|---|---|---|
14 | Can a clash-free exam timetable use fewer slots than your school's? | Graph colouring, chromatic number | Anonymised subject-choice data |
15 | How does key length change the time to break a small RSA key? | Modular arithmetic, primes, factorisation | Your own code, timed runs |
Idea 14 has an unusual advantage: if it works, your school can use it.
Grades 6 to 8 can scale these down rather than swap them out; ideas 3, 4 and 7 work with a smaller sample and a simpler test, roughly the level of the Young Scholar Program.
How to write the introduction of a mathematics project?
This is the section students rewrite most and plan least. A mathematics project introduction is not a definition of the subject. In 120 to 180 words, it answers four things: what exactly you are asking, why it is worth asking, which specific tools you will use, and what form your answer will take. Name the tools, not the branch; "chi-square goodness of fit" tells a reader something, "statistics" does not. Applied to idea 3:
Dice are assumed to be fair, but manufacturing tolerance and wear can bias them. This project tests whether the six dice in a standard board-game set produce outcomes consistent with a uniform distribution. Each die is rolled 600 times under identical conditions, and observed frequencies are compared with the expected 100 per face using a chi-square goodness-of-fit test at the 5% level. The report states, for each die, whether the deviation is larger than chance would comfortably explain.
Write it last, once you know what you found. It then takes twenty minutes rather than three evenings.
How to write the conclusion of a mathematics project?
The conclusion of a mathematics project is not a summary of what you did; that belongs in the abstract. It states what is now known that was not known before, in four parts.
The answer, numerically. Not "the dice were mostly fair" but "five of six dice gave p-values above 0.05; die four gave 0.012."
What it means. A sentence placing that number back in the original question.
How confident you are. Sample size, assumptions, anything that could have distorted the result. Naming a weakness makes a project stronger: it shows you understood the method.
What you would do next. The version you would run with more time or better instruments.
If your conclusion could be pasted into another student's project on the same topic unchanged, it is not a conclusion yet.
Four mistakes that weaken math projects
Collecting data and then only averaging it. A mean and a bar chart are descriptions. It becomes mathematics when you test the data against a model.
Restating a known proof and calling it research. Reproving that √2 is irrational is a fine exercise and a weak project. Apply it, extend it, or test where it breaks.
Sample sizes too small for the test. Chi-square behaves poorly when expected counts per category fall below 5. Determine your sample size for the test you intend to use before collecting anything.
Measurements reported without error. A height quoted as 24.7 metres from a phone app implies precision the method lacks. Give a range, and say how you got it.
What does a math project have to do with career opportunities in mathematics?
The career opportunities in mathematics people actually enter- actuarial work, data science, operations research, quantitative finance, cryptography, biostatistics, teaching, and academic research- share one daily habit. Someone hands you a messy real situation, you decide which model applies, then defend the choice.
A school project is the earliest place to practise that, and it is why admissions readers respond to project work. A student who chose a method, hit a problem in their data and wrote honestly about the consequence has shown something a transcript cannot. No single project decides an application, but the habit compounds.
A realistic timeline
Stage | Time | Where students lose weeks |
|---|---|---|
Narrowing the question and reading around it | 2–4 weeks | Stopping at a topic, not a question |
Collecting data | 3–5 weeks | No pilot run, so the format changes halfway |
Analysis and writing | 3–4 weeks | Choosing a test you cannot interpret |
Review and revision | 2–3 weeks | Leaving the introduction and conclusion till last |
Sixteen weeks from question to finished draft is normal for tier two. Publication sits on top of that, not inside it, so check what a target journal accepts before designing the study; our list of student-friendly journals sets out what each takes.
Narrowing is where guidance changes the outcome most. An experienced researcher will tell you in one conversation that your sample is too small for the test you picked, then suggest the version that works. That is what the first four weeks of the High School Scholar Program are built around, and our research project ideas carry worked examples in other subjects.
Frequently asked questions
Can a mathematics project be done without collecting any data? Yes. Ideas 7, 14 and 15 run on proof, code or public information alone, provided the output goes beyond restating existing results.
How long should the introduction and conclusion be? Roughly 120 to 180 words each for a school project. Both are short sections doing precise jobs; length is not what makes them good.
Is an inconclusive result a problem? Not if the method was sound. "The facades I measured showed no clustering near 1.618" is a finding, and reporting it rather than adjusting the data to fit is what makes it research.
