Protective experiments are a statistically robust way to validate equivalence when there is no expectation that a product change improves any of our key metrics. It validates our ability to maintain the status quo.