I'm suspecting that the projected MJO is a factor causing the GEFS to be on its own. The MJO forecasts are unusually far different for the next few days with the GEFS staying around where it is inside the circle near phase 2 followed by an unusual clockwise rotation backwards into phase 1 whereas the usually more reliable EPS has it rotating counterclockwise into phase 3 followed by phase 4 pretty quickly. Usually the EPS verifies more closely fwiw. With the GEFS clockwise rotation being unusual vs climo and with it usually not doing well when far different from the EPS, I certainly wouldn't bet on its MJO forecast right now. It will be interesting to see which model comes closer. These model differences make following TCs more interesting to me.