The leverage is a numerical value between 0 and 1 that indicates the potential for a design point to influence the model fit.
A value of 1 means that the model will exactly fit the observation.
Limits: A run with leverage greater than 2 times the average leverage or has a leverage of 1 has high leverage.
A high leverage point is only “bad” if there is a problem with that data point (unexpected error). However, if the data is a good representation of how the process behaves at that setting, then it does not invalidate the model.
Smaller designs will generally exhibit more leverage for individual points. A run in a large design will exhibit high leverage when it has factor settings that are not similar to other runs. High leverage also occurs when the number of terms in the model approaches the number of runs.
Never ignore a run just because the diagnostics plots indicate it may be a problem. Verify that the data is wrong in some way before ignoring it.