Predicting in decision rules

Sequential covering is a type of decision rule procedure that repeatedly learns a single rule to create a decision list (or set) that covers the entire dataset rule by rule.

Given a training dataset, we can find a set of rules that cover the whole data.

When it’s time to predict, one goes through the rules till it finds one that fits the individual prediction.

What happens if there is no perfect rule for the individual prediction?
Does it predict the average of the class?