Quantum computing (applying quantum gates to quantum states) is so hot right now, but how good are your gates? Randomized benchmarking is an experimental technique aimed at estimating the average error in a set of gates. Due to it's ease of implementation and propensity to produce favorable numbers, it is now the de facto standard. However, standard practice can lead to the wrong conclusions. Here I will use a simple analogy (coin flipping, surprise!) to convince you that one seemingly innocuous choice (sequence reuse) is not only problematic, but a symptom of a bigger problem.
(PLEASE NOTE NON-STANDARD DATE)