The Model dillema: Rain prospects and model reliability

For California the weather the past 8 days or so has been fabulous with warm sunny days and pleasantly cool nights, however the models are forecasting a change to a wetter/cooler pattern for California around the 1st of next month. Even though the change is forecast by some models the details are far from certain. With each successive run of the GFS bringing less rain to both NorCal and SoCal, it goes without saying that the "devil is in the details." While prospects of a strong storm for NorCal looked promising on the GFS about two days ago, now the model has been less rigid in its forecast and forecasts a meager 1/2 inch for Napa on the 1st when compared to the two to three inches forecast just a couple of days ago. This of course brings up the issue of model reliability and when is a forecast close enough to the day it's forecasting that we consider it to be "reliable enough?" The issue is more personal preference then anything else but for me I'll stick to 5 days as the optimal period of reliability from the models. So for the forecast for next week, since it's still 10 days out, the best approach is a subtle decision to wait and see, rather than to jump the bandwagon of a wet forecast. At this point in time the teleconnections show the possibility of a quick hitting trough followed by strong ridging with temps once again reaching the 90's for inland areas, so even if we see a change it will be brief. As it is with the change in seasons, it's two steps forward and one step back.