Yes the bullet does but it also has a way higher velocity, it's path decays so quickly as the water put's up an opposing force (Newton's 3rd law of motion).
But you are neglecting the movement of the water itself. When the pump first turns on, there is a lot of resistance but as time goes on and the pool currents are setup, the resistive force is reduced and the directional flow moves further and further away from the return reaching farther areas of the pool.

Also, I didn't mean to imply that the currents are only in one straight direction. I was mainly referring the fact that water does not drop like the bullet does. I understand that the water travels more or less in a cone and the cone angle is dependent upon the orifice size and shape.


except the single lines for the returns is way too much
You mentioned that you try to keep the port size the same between the suction and return sides but the above quote would indicate that you do not bring those same ports sizes (i.e diameter) all the way to the pump for at least the return side. If that is the case, the return side would definitely have more head loss due to less parallel pipe paths than the suction side. So I have to ask, isn't that defeating the purpose of the whole "port size" objective? Pipe, fittings and the exit orifice all add head loss to varying degrees so I don't see any benefit having the port sizes equal other than they have slightly less head loss than a smaller orifice eyeball. Is that the main objective?

Also, I am curious about your pump. Can you give more details about it (make/model)?