Bear Catches the 5th Fish

This quant question is about a simple sequential random process with a stopping rule, wrapped in a story so it feels intuitive during interviews. It lives firmly in basic conditional probability and discrete distributions, but forces you to think about what it actually means to "stop" a process and how that affects later events. Even though the setup looks like an easy brainteaser, it hides a very standard pattern used all over quant finance.

It trains your ability to reason about paths of a stochastic process when there is a target number of successes. This is core quant prep: reading the problem carefully, formalizing the random experiment, identifying relevant scenarios, and assigning the correct conditional probabilities. It also checks whether you can keep track of dependence introduced by a stopping condition in an otherwise independent sequence.

This matters for quant interviews because many trading, risk, and execution problems involve hitting thresholds, barriers, or targets. Interviewers want to see whether you can quickly translate a verbal story into a probabilistic model and analyze it rigorously under time pressure. Being fluent with such stopping-time style questions signals strong foundations for more advanced Markov, martingale, and optimal stopping problems that appear in real quantitative finance work.

What it tests

This class of problems is governed by the principle of conditioning on stopping times in sequential probabilistic processes. When an agent stops after achieving a certain number of successes, the probability of any event that could occur after each attempt depends on whether the stopping condition has been met. The key is to partition the sample space into mutually exclusive scenarios based on whether the process has stopped or is still ongoing at a given step. This allows you to use the law of total probability, weighting each scenario by its likelihood and the conditional probability of the event of interest within that scenario. The structure arises because the process is memoryless and each trial is independent, but the stopping rule introduces a dependency on the history of outcomes up to each point.