This presentation puts the method of sequence analysis by optimal-matching into a broader context by focusing on differences in analysing (learning) paths by markov-chains. These methods of process analysis have different perspectives: While sequence analysis focuses on patterns of processes, markov-chains focuses on probabilities of dyadic transition from one step to the next in a process. While the result of sequence analysis as a distance-matrix presents a typical pattern of steps or phases (i.e.: o-e-a-r-t) the result of markov analysis as a probability-matrix are sequences of probabilities that characterizes the transitions between steps within a process pattern (i.e.: p1-p2-p3-p4). Combining both sequences one gets a mixed sequence like the following: o-p1-e-p2-a-p3-r-p4-t. Sequence analysis by optimal matching uses Levenshtein distance as a mesure for the simularity of sequences. In my presentation I will show how to include the probabilities as parameters in this mesure of simularity.