in some log L. Now consider a model that cannot replay σ ,
but that can replay 99 of the 100 events in σ (i.e., the trace is “almost” fitting). Also
consider another model that can only replay 10 of the 100 events in σ (i.e., the trace
is not fitting at all). Using the naïve fitness metric, the trace would simply be classified as nonfitting for both models without acknowledging that σ was almost fitting
in one model and in complete disagreement with the other model. Therefore, we use
a fitness notion defined at the level of events rather than full traces.