In this talk, I will compare the discriminatory power of the cut-and-count method with the matrix element method, in constraining a simple extension of the SM, namely the Lmu-Ltau interactions. The Z' associated with the spontaneously broken U(1)_{Lmu-Ltau} symmetry only interacts with the second and third generation of leptons at tree level, and is thus difficult to produce at the LHC. I argue the best channels to look for such Z' are Z' -> 4mu and Z'->2mu+ MET. Both of these channels have a large number of observables which strongly motivates the usage of a multivariant technique. I will show that the matrix element method as a multivariant technique - with some modifications to make the calculation easier- can improve our sensitivity by a factor of 5 to 10 compared with the cut-and count method.
