You can have the largest HPOP in the world, but if the delivery system (fittings, lines, rails, etc.) bottleneck the output, all you are going to end up with is a tremendous pressure drop at the injector as soon as the spool valve opens. There is a reason why Ford moved the ICP sensor directly to the top of the oil rail. This was to provide an accurate pressure reading at the injector.
For the sake of argument, let's break it down to a real-world analogy that anyone can try... A garden hose. Our system comparison is as follows:
The water main to the house is your HPOP.
The spigot (valve) is the HPO system restriction. (HPO lines, fittings, and overall pressure loss from distance.)
The garden hose is the HPO rail in the head.
The spray nozzle is the injector.
The opening of the spray nozzle is the injection pulsewidth.
Take the garden hose with a spray nozzle on the end of it and make sure the nozzle is closed. Crack the spigot a tiny bit and let the pressure in the hose build up. Now open the spray nozzle and you get a quick burst at high pressure but it then immediately peters out to little more than a dribble. Close the spray valve and pressure builds back up again. If you cycle the opening and closing of the spray nozzle correctly (small, controlled pulsewidth), you will always get a good burst of water when opening the nozzle. Hold the nozzle open too long (large pulsewidth) and it will peter out. Remove the restriction (open the spigot) and you can hold the nozzle open all day long with no drop in pressure at nozzle.
Yes, this is probably an oversimplification, but it doesn't make it any less relevant.
I've looked at the HPODS that is offered on DieselSite and their solution is to add a second point of entry for oil into the HPO rail by utilizing the ICP sensor ports of each head. (Modification required, of course!) Since this is a parallel system, this effectively doubles the delivery (flow) capability between the HPOP and the HPO rails.
If you aren't convinced there is a restriction or pressure drop in the system, take an ICP sensor and install it in the pump cover (like on an '03) and then measure the difference between the pump mounted sensor and the head mounted sensor. Any difference in pressure will clearly indicate a FLOW problem between the HPOP and the injector rails.
In fact, I will be doing this EXACT test next week before I install the injectors. I will be mounting a second ICP sensor at the pump and then taking readings from both sensors running different programs. I'll post up the datalogs and then we can settle this debate entirely.