Blue After First Blue Draw

This quant interview question is about how new information changes probabilities in a simple random experiment. You start with symmetric-looking choices and then observe an outcome that reveals partial information about which world you are in. It sits at the intersection of conditional probability and Bayesian thinking, but wrapped in a clean, concrete setup that shows how intuition can be misleading if you do not update correctly.

It trains comfort with conditional probability, the law of total probability, and Bayesian updating in a setting with hidden states. Good quant prep means getting used to translating "I observed this" into a new probability distribution before answering anything about the future. It also trains clear notation, careful conditioning, and the habit of checking whether you are using prior or posterior information.

This matters for quant interviews because real trading and risk problems look exactly like this in disguise. You rarely know the true state of the world; you only see noisy signals and must constantly update your beliefs. Interviewers use questions like this to see whether candidates can reason cleanly under uncertainty, avoid common conditional probability traps, and communicate a logically consistent argument under time pressure.

What it tests

This problem class is governed by the law of total probability and conditional probability, especially as applied to Bayesian updating. When you observe an outcome that provides information about an underlying hidden state (like which urn was chosen), you must update your beliefs about that state using Bayes' theorem. The key is that the probability of future events depends on this updated belief, not just the original setup. This structure appears whenever you have a random selection from several possible sources, and then observe evidence that changes the likelihood of each source. The principle holds because each observation acts as a filter, reshaping the probability distribution over the hidden variables that drive the process.