Thanks Matt!  You hit the nail on the head that there tend to be different *drivers* (influencers) for different averaging periods.  For example, one wouldn't lean on the polar vortex or MJO to make a seasonal forecast.  There might be a few rare exceptions, say if the MJO was already in the *initial conditions* of a model run and was conducive to strong storms in a typically arid location.  But, mostly, we do not know what the state of the SPV/AO/MJO/etc will be during the course of an entire season (90 day average) and it's also possible that, within the course of a season, changes in those phenomenon could cancel themselves out (The Arctic Oscillation/AO could be positive for the first half the season and then be canceled out by AO negative during the second half).  So that's why we lean on factors like ENSO, trends, and other surface boundary conditions when we make a *seasonal* outlook.