Interviewers quote you odds and watch what you do. The skill being tested is not arithmetic — it is whether you can tell a generous price from a poor one, and whether you know how much to commit when you are right.
Start the Probability Pricing drillFree, runs in your browser, nothing to installFractional odds of b : 1 carry an implied probability of 1 / (b + 1). Odds of 3:1 imply 25%. Odds of 1:1 imply 50%.
So every quoted price is a claim about how likely something is. Your job is to work out the true probability and compare. If the true chance is 30% and the price implies 25%, the quote is generous and you take it. If the true chance is 20%, you pass — and passing costs nothing.
Most candidates lose points not by miscalculating, but by betting on everything. Taking a bad price is worse than taking no price.
Knowing you have an edge is half the answer. The other half is size. The Kelly criterion gives the fraction of your bankroll that maximises long-run growth:
f* = (b·p − q) / b
where b is the odds, p the true probability, and q is 1 − p. At even odds with a 60% chance, f* = 0.20 — you commit a fifth of your stack, not all of it.
This matters to a trading firm because it is the difference between a trader who compounds and one who eventually goes to zero. Betting more than Kelly increases risk while reducing long-run return, which is the counter-intuitive result they want to see you understand.
Each round also quotes two side markets: Insurance pays if your main positions lose money, Boost pays if they make money. They look like novelties. They are not.
When Insurance is priced generously and you also hold a strong position, the two together can produce a profit in every possible outcome — a genuine arbitrage. The drill detects this exactly, by enumerating the entire sample space rather than estimating, and awards a bonus when you find one.
Spotting that structure is precisely the instinct market makers are hired for.
After each round you see the true probability, the implied probability, your edge, the Kelly-optimal stake, and whether each call was right. The score separates decision — did you take the good prices and pass the bad — from sizing — how close your stake was to Kelly.
Those are different skills and they fail differently. Track which one is costing you.