Is the pattern in the evidence or in the way the data was processed?
Reproduce important calculations and links, record uncertainty and test reasonable alternative explanations. Separate correlation from causation and assess whether linked records truly identify the same person, event or device.

