Test flakiness rate
The proportion of tests that produce inconsistent results, passing and failing on the same code without changes, usually from timing, shared state, or environment assumptions. A high flakiness rate erodes trust in the whole suite: once a red build is routinely dismissed as 'probably just flaky', real failures slip through unnoticed.
Flaky tests, ones that pass and fail on the same code, are corrosive because they teach the team to ignore red, and an ignored signal is worse than no signal. Tracking a flakiness rate makes the problem visible and fundable instead of a thing everyone grumbles about and re-runs. A healthy team quarantines flaky tests fast and fixes the root cause, timing, shared state, environment drift, rather than normalising "just hit retry."