Stating Assumptions and Calibrated Confidence

Two candidates both answer "about 200,000" to a market-sizing question, but one says it flatly while the other says "I'm estimating around 200,000, and the piece I'm least sure about is the average purchase frequency — if that's off by half, this could easily be 100,000 to 400,000." The second answer is far more useful to an interviewer, because it shows the candidate knows where their own estimate is shaky, which is a skill traders use constantly when sizing a position around an uncertain number.

Say the assumption, not just the number

Every estimate rests on assumptions, and naming them out loud turns a guess into a piece of reasoning the interviewer can evaluate. Instead of "there are about 5 million cars in this city," say "assuming a population of about 10 million and roughly one car per two people, that's 5 million cars." If the assumption is wrong, the interviewer can correct it and you adjust the estimate on the spot — which looks far better than defending a number pulled from nowhere.

Say how confident you are, and why

Calibration means your stated confidence should track how much the answer would move if a key assumption changed. If your estimate barely depends on an uncertain input, say so — "even if that assumption is off by a lot, the final number barely changes." If it depends heavily on one shaky number, flag that too: "this whole estimate really hinges on the tuning frequency, which I'm least sure about." A candidate who states every number with the same flat confidence, whether it's solid or shaky, is missing a skill that matters directly on a trading desk — knowing which part of your view to trust and which part to hedge.