Even flips till repeat

This quant interview question is about probabilities in sequences of independent random events, framed in a very simple setting that hides a deeper structure. It focuses on how a stopping condition based on patterns changes the distribution of the total length of the process. Despite the basic context, it is a classic example of nontrivial behavior emerging from a seemingly trivial game.

It trains your ability to model stopping times, recognize rigid pre-terminal structures in paths, and connect them to distributions over path lengths. Working through it builds intuition for pattern-avoidance in stochastic processes, conditioning on partial histories, and relating combinatorial structure to probability weights across different stopping scenarios.

This matters for quant interviews because it mirrors how you analyze path-dependent products, regime switches, and barrier-like events in markets. Top trading firms use such questions to test whether your quant prep includes comfort with Markov-style reasoning, discrete-time models, and clean, rigorous probabilistic thinking under time-dependent rules.

What it tests

Problems involving the first occurrence of a specific pattern in a sequence of independent trials often reduce to analyzing the structure of sequences that avoid the pattern until the final step. The key is that, to delay the pattern, the sequence must follow a rigid structure (such as alternation) up to the penultimate step. This creates a one-to-one correspondence between the length of the sequence and the number of ways to arrange the outcomes, often resulting in a geometric series when summing probabilities over all possible stopping times. The reason this works is that the constraints force all but the last few outcomes to be uniquely determined, making the counting tractable and the probability calculation systematic. This approach generalizes to many 'first occurrence' problems, where the avoidance of a pattern imposes a deterministic structure on the sequence.