Do you properly flag Trusk submissions? Otherwise the simulation won't be very accurate at all.