“To prove it I booted Linux 292,612 times before the faulty commit (successfully), and then after (failed after under 1,000 boots).”
Linux 6.4 hangs on roughly 1 in 1,000 boots, and nobody noticed until a libguestfs test suite kept stalling at the same spot. Finding it took days of bisecting between v6.0 and v6.4-rc6, with 10,000 boots to clear each step. The culprit was a regression in the printk time feature. A one-in-a-thousand bug gets blamed on flaky hardware, and only a paranoid test harness catches it.
