A school quiz settles a question with a buzzer and a right-or-wrong answer. A coding contest settles a problem against test cases a program either passes or doesn’t. A school or college math olympiad settles neither way: most of its marks come from a human judge reading a written proof or a multi-step working and deciding how much of the method is correct, even when the final numeric answer is wrong — and deciding how much a right answer is worth when the working shown doesn’t support it. That judged, partial-credit, proof-based shape is what makes a math olympiad a genuinely different scoring problem from every other “intelligence item” on a fest programme, and it’s why organisers who try to mark one like a quiz or an auto-checked coding round usually run into disputes at results time.
What actually gets scored in a math olympiad
School and college-level math olympiads — whether run as a standalone “Math Olympiad” stage item, a math-fair contest alongside project exhibits, or a college department’s annual meet — settle into a small number of recognisable formats:
- Objective round: a fixed set of problems with a single correct numeric or multiple-choice answer each, auto-markable like a quiz — the easiest format to run, but it only tests the final answer, not the method.
- Subjective / proof round: a small number of problems (often 3–6) each marked out of a fixed total (commonly 7 or 10 per problem, the pattern most national and international olympiads publish), where a marker reads the full written solution and awards partial marks for correct steps, valid lemmas, or a right approach that stops short of a complete proof.
- Timed speed round: many short-answer problems against the clock, scored purely on count correct within the time limit — closer to a rapid-fire quiz than a proof round, often used as a school-level qualifier before a subjective final.
Most real-world school/college olympiads actually combine two of these — an objective or speed qualifier to shortlist, followed by a subjective proof round for the final ranking — so a rulebook should say up front which stage a given event is, since the scoring software, the markers needed and the time per problem differ sharply between them.
Partial credit on a proof: the hardest mark to print in advance
Unlike a quiz where an answer is simply right or wrong, a proof-based problem is almost always partially solved: a student may set up the correct equation but make an arithmetic slip, or prove most of a result but miss the final step. Published marking schemes handle this in one of two ways:
- Step/rubric marking: the setter publishes (or the marking panel agrees beforehand) which intermediate steps earn how many of the problem’s marks — e.g. 2 marks for the correct setup, 3 for the main working, 2 for the final step — so two independent markers reach the same score for the same script. This is the standard used by national-level olympiads and is the only approach that scales to more than one marker without disputes.
- Holistic judgement: a single marker reads the whole solution and assigns a mark out of the total based on overall correctness and rigour, with no published step breakdown — faster to run with one marker and a small field, but it does not survive a second marker re-checking the same script and gives organisers nothing to show a parent who disputes a mark.
A right final answer with no working shown, and a wrong final answer with mostly-correct working, are the two cases every marking scheme should state a rule for before the papers are collected — most published schemes award only partial credit for the first (a correct answer alone is not a proof) and meaningful partial credit for the second (the method, not just the number, is what’s being tested).
The tie-break order that actually gets used
Because marks-out-of-a-fixed-total produce frequent ties, especially among the top few students at any one level, a clear, printed tie-break order matters as much here as the marking scheme itself. A dependable published order is:
- Higher total score across all problems
- Higher score on the problem (or problems) the setter has pre-designated as the “decider” — typically the hardest problem on the paper
- Fewer total problems attempted with zero marks (rewards a student who attempted fewer problems but solved them more completely)
- A pre-announced tie-break paper or oral round, used only at the very top of a close field (state/national-level events)
A rulebook that stops at step 1 is the single most common cause of an olympiad results dispute — organisers should decide and print this order before the paper is set, not after the scripts are already marked.
What belongs on the registration form, not the marking scheme
A few decisions genuinely belong to event setup rather than scoring, and a rulebook that keeps the two separate avoids confusing the volunteers running registration and the panel marking scripts:
- Age-group/grade bands for the paper set (most olympiads print a different, harder paper for higher grade bands rather than one paper for everyone)
- Calculator policy (allowed/banned, and whether it differs by round — common for the objective round to ban calculators and the subjective round to allow them)
- Time per round (objective rounds commonly run 45–60 minutes; a subjective round with 3–6 problems commonly runs 90–180 minutes)
- Number of markers per script and whether a second-marker check applies to borderline or top scripts
- Whether results are published as marks, ranks, or grade bands only (some olympiads deliberately publish only a grade/band, not the raw score, to avoid comparing marks across different paper sets)
Setting this up on eTalenter
A math olympiad fits eTalenter the same way the site’s other scoring-mechanics items do — chess, sudoku, typing, coding and robotics all follow this pattern:
- Add it as an Individual competition item under the relevant age Group/Level, the same place a quiz or a coding item sits.
- Run the paper-setting, invigilation and marking itself outside eTalenter (on paper, or whatever marking panel/rubric the format needs) and bring only the final, already-decided marks or ranks into eTalenter.
- Use Assign Results Manually to enter each participant’s final rank (or score, if the event wants marks published) once the tie-break rule above has already settled who placed where — eTalenter doesn’t mark proofs or re-check scripts itself, the same division of labour it uses for a chess result or a coding contest’s final standings.
- Print certificates and the results sheet straight from the event once results are in, so the olympiad item flows through the same results cascade as every other competition on the programme.
Frequently asked questions
Can a math olympiad be marked the same way as a quiz?
Only the objective round, where every problem has a single correct answer. A subjective/proof round needs a published step-marking rubric, because partial credit for partly-correct working is the whole point of the round.
What’s the fairest way to award partial marks on a proof?
Publish (or agree among markers before marking starts) how many of a problem’s marks go to each intermediate step, so two different markers would give the same script the same score. A holistic “read it and judge” mark works for a small one-marker event but doesn’t survive a second-marker check.
What’s the fairest tie-break for an olympiad?
Total score first, then the score on a pre-designated decider problem, then fewest zero-mark attempts. Print this order before the paper is set, not after marking.
Does a right final answer always get full marks?
Most published schemes say no — a correct number with no supporting working is not a complete proof and earns only partial credit; decide and print this rule in advance rather than leaving it to the marker on the day.
Does eTalenter mark the papers itself?
No. eTalenter records results, ranks and certificates the same way it does for chess, sudoku or a coding contest — the paper-setting and marking panel (however the organiser runs it) decides the marks; eTalenter is where the final result gets recorded, published and certified.
Related reading
See also eTalenter’s guides to coding competition judging, sudoku competition judging, and Kerala school maths fair item codes for the same results-first scoring pattern used across eTalenter’s intelligence-item competitions.
Run your math olympiad on eTalenter
eTalenter handles registration, grouping by age/grade, results entry and certificates for a math olympiad the same way it does for every other competition item on a school or college fest. See it running on a real event: open a live demo and log in as a demo coordinator or judge — no sign-up needed. Explore the full feature set on the kalolsavam management software page, message us on WhatsApp, or write to support@etalenter.com. We help with the eTalenter software itself — not with your own competition entries.
Was this guide useful?
