PRACTICE GUIDE OPTIVER
Optiver Beat the Odds Practice Test
The Optiver beat the odds screen is 20 questions in 14 minutes - about 42 seconds each - answered by typing the number, with no calculator. It marks negatively, so a wrong answer costs you a mark and a skip costs you nothing.
Where it sits: Section D of the same battery. The practice sitting on this page runs the same item count, the same clock and the same marking rule, with questions generated by Outcry rather than taken from Optiver.
Outcry is not affiliated with Optiver and has no access to their assessment content. This guide describes an assessment format that candidates report publicly; the questions here are generated by Outcry and are not Optiver’s own.
What it screens
A global market maker headquartered in Amsterdam, trading options, ETFs and futures on major exchanges.
- ✓Expected value computed fast and used to make a decision
- ✓Conditional probability and Bayes updates without formal notation
- ✓Ranking likelihoods rather than computing them exactly
- ✓Fermi estimation: decomposing an impossible question into three easy ones
Where it sits at Optiver
This section does not arrive on its own. It is section D of 6, and the sitting around it runs 95 minutes and 175 items in total, with 80 in 8, NumberLogic, Likelihood List, Zap-N: number memory and Zap-N: patterns on the same continuous clock. Candidates who prepare only for this round tend to be caught by the pacing of the rest.
There is no going back. You answer or skip, and the next item loads, which means the decision to leave an item is final at the moment you make it.
The format
These are the numbers the Optiver sitting on this site runs on, matching the format candidates report.
| Questions | 20 |
|---|---|
| Time | 14 minutes |
| Per question | 42 seconds |
| Negative marking | Yes - a wrong answer costs a mark |
| Answer style | Typed numeric answer, no calculator |
| Where it sits | Section D of the same battery |
What it tests, with a worked example
Every example below is generated by Outcry, drawn from the same question generators the timed drills run. None of them is Optiver’s.
Expected value and odds
Pricing a bet, converting between odds and probabilities, and deciding whether to take it.
Example
Roll 2 fair dice. Probability every die shows a six?
- 1/216
- 1/36
- 2/6
- 1/12
Answer 1/36
Independent rolls multiply: (1/6)^2 = 1/36.
Conditional probability
Bayes' rule in disguise: a test result, a signal, a drawn card, and what it implies.
Example
You have one fair six-sided die and need a uniform random number from 1 to 7. Which procedure works?
- Roll twice and take the larger, mod 7, plus 1
- Roll twice and take the sum mod 7, plus 1
- Roll once and multiply by 7/6, rounding
- Roll twice for 36 outcomes; re-roll on the last one, map the other 35 into 7 groups of 5
Answer Roll twice for 36 outcomes; re-roll on the last one, map the other 35 into 7 groups of 5
36 outcomes can't split into 7 equal groups, but 35 can (7×5). Discard one outcome and re-roll - the sum-mod-7 trick fails because sums are not uniform.
Distributions you should know cold
Binomial, geometric and uniform means and variances, applied rather than quoted.
Example
Two fair 6-sided dice are rolled out of sight. At least one shows a 6. What is the chance both do?
- 18.2%
- 2.8%
- 16.7%
- 9.1%
- 50%
Answer 9.1%
Outcomes with at least one 6 number 11, and exactly one of those has both, so 1/11 = 9.1%. The intuitive 1/6 answers the different question where a NAMED die shows a 6.
Waiting times and repeated trials
How long until the first success, and why the answer is rarely the average you expected.
Example
A fair coin is flipped 40,000 times. Give the number of heads at the TOP of the central 95% range. Nearest integer.
Answer 20196
The count has mean 20,000 and standard deviation √40,000/2 = 100. Two of those above the mean: 20,000 + 2 × 100 = 20,200.
Symmetry arguments
Questions that collapse to one line once you notice the two cases are mirror images.
Example
An index of 500 names is replicated with the largest 80, leaving 420 names worth 3.6% of the index unheld. Those 420 are equally weighted within that 3.6%, each carries 40% of idiosyncratic volatility, and their idiosyncratic moves are independent. What is the annual tracking error of the basket?
- 1.440%
- 40.000%
- 0.070%
- 0.003%
- 3.600%
Answer 0.070%
The systematic part of the missing names is already carried by the 80 held, so what is left is 420 independent idiosyncratic bets. Each missing name has weight 3.6%/420 = 0.0086%, so on its own it contributes 0.0086% x 40% = 0.0034% of volatility. Independent risks add in quadrature, not in a straight line, so 420 of them come to sqrt(420) x 0.0034% = 20.494 x 0.0034% = 0.070%. Adding the risks straight rather than in quadrature gives 1.440%, which is what you would get if all 420 missing names moved together, and it is sqrt(420) = 20.494 times too large. The residual is small precisely because the omitted names are many, small and unrelated, which is the whole case for partial replication.
Reading the question exactly
At least, exactly, and given that - the three phrases that change the answer entirely.
Example
Revenue by division, $m Division 2016 2017 2018 2019 Rates 140 420 430 325 Macro 390 325 335 300 Credit 305 160 230 480 Equities 175 195 155 225
What was Rates average annual revenue across the four years shown?
- 1315
- 285
- 328.75
- 325
Answer 328.75
140 + 420 + 430 + 325 = 1315, over four years that is 328.75. The midpoint of the highest and lowest, 285, is not the mean.
What a good score looks like
On a paper of 20 questions marked +1 and -1, the arithmetic of the threshold matters as much as the threshold: every wrong answer costs you two places against someone who skipped it. Probability rounds are usually marked as part of a wider battery, so a standalone threshold is rarely reported. Where candidates do describe one it clusters around getting two thirds of the items right, with the caveat that these sections are short enough that two mistakes move you a long way.
How to train for it
- 01Work the standard results until they are recall, not derivation: complement, conditional, binomial mean and variance, geometric expectation.
- 02Do them timed. A probability question you can solve in four minutes and the paper gives you ninety seconds for is a question you cannot solve.
- 03Work the complement first whenever the question says at least or at most. It is usually the shorter route and it is almost always the intended one.
TRAIN IT HERE
The drills that match each section
Likelihood Ranking
Order events from most to least likely - the exact screen-question format.
Fermi Estimation
Piano tuners in Chicago, atoms in a body - scored on order of magnitude.
Quitters Never Lose
A casino stopping-strategy game that is secretly a martingale lesson.
Tail Risk
Rare events and fat tails, and what they do to a plain expected-value estimate.
SIT THE FULL BATTERY
All the sections back to back on one clock, marked the way the real screen marks them, with a by-skill breakdown at the end. Included with any pass.
Also reported at Optiver
Common questions
- Is the Optiver beat the odds test multiple choice?
- Typed numeric answer, no calculator. You type the number, so there is nothing to eliminate your way to.
- How long is the Optiver beat the odds test?
- 20 questions in 14 minutes, which is about 42 seconds each.
- Is there negative marking on the Optiver beat the odds test?
- Yes. A wrong answer costs you a mark, so an item you cannot see a route through is worth skipping outright - a blind guess on this paper is negative expected value.
- How do I practise for it free?
- Every drill linked on this page is free to play, with no account, inside a daily run cap. Questions are generated fresh each run, so there is nothing to memorise between attempts. The full Optiver Trading Assessment sitting puts the sections back to back on one clock.