Bad Qubits — Learn Quantum by Doing

Multiple choice

In minimum-weight perfect matching (MWPM) decoding under independent noise where each qubit errs with small probability p < 1/2, why does choosing the lowest-weight error consistent with the syndrome maximize the success probability?

Because the lowest-weight error is always the unique error producing that syndrome
Because P(config) is proportional to (p/(1-p))^w with p/(1-p) < 1, so probability decreases monotonically with weight w
Because higher-weight errors are physically impossible when p < 1/2
Because matching weight is unrelated to probability, and minimizing it is chosen only for computational speed