Run variation sources the axes along which ceteris paribus can be violated

Each of the axes, when changed between experiment runs without intent could produce non-deterministic results. Deliberately changing one of those per run allows the experiment to clearly show the impact of that single change while respecting ceteris paribus rule.

Those axes are:

• data
• logic
• configuration
• randomness
• environment

see The replication-variance dimension classifies experiments by whether they produce a noise distribution for the associated experiment dimension see A replication knob is a source of variation under a fixed experimental design for an example of randomness axis