In the typical experimental protocol, the maximum skip for each primary ELS is set by requiring that the expected number of primary ELSs for each key word is a user specified constant, like 10 or 30. As this expected number of ELSs increases, the probability increases that there will be some chance meeting of ELSs that will constitute a minimal or near minimal geometric configuration. Thus for large expected numbers of ELSs there will almost surely be monkey texts with maximally compact or near maximally compact geometric configurations of ELSs. These will tend to increase the p-level of Torah text even if its ELSs are in near maximally compact configurations. Once the probability of maximally or near maximally compact configurations of ELSs for monkey texts gets greater than 1/1000, then p-level for the Torah text must be greater. This effect is called the saturation effect .
This is the reason that the Torah code hypothesis specifies that the primary ELSs must have small skip rank. When primary ELSs have small skip rank, the saturation effect is held to a non-significant value.