I've been thinking about that, too -- a lot.
Welcome to the 'practice" of medicine.
As just an FYI between 2000 and 2005 the AHA reported cardiac arrest mortality increased by 50,000 over the previous guidlines. (which focused heavily on medication and other "advanced" procedures.
Simplfying the 2005 guidlines to get back to basic CPR has seen a dramatic increase in survival, (the actual numbers escape me, I will look them up when I am at home.)
SO how many people died before we figured out all those "advanced life saving" measures didn't matter? About 50K.
If you consider the actual physiology behind oxygenation, you find a couple of interesting things. First and foremost, the human body only requires 11% oxygen to survive. (the same amount for an open flame for the fire folks)
Under normal circumstances, each RBC only gives up 1/4th of its heme saturation. So there is a considerable reserve. (though I admit it is not so simple, and it would take me pages to describe here, get a physio text)
Adding large quantities of oxygen after a period of hypoxia is showing to trigger mitochondrial apoptosis cascades. Furthermore, there is a vasoconstrictive property to oxygen.
for years, large volume oxygen therapy has been a mainstay of EMS. But despite the fact it is classified as a drug, somehow there became an idea if some is good, more is better. Most medical, nursing, and educated allied health professionals have been using more precise focused oxygen therapies. which is not to say that maximum oxygen is always detrimental, it has its place, but it is not every patient.
As an interesting anecdote, I was dicsussing this very topic in an ACLS EP course last year, all the hospital staff in the specific scenario called for oxygen by cannula. All the EMS people asked for NRB. When asked to explain the reaosn for cannula, a very knowledgable reply about enzyme kinetics was offered. When asked the Same for NRB, the answer was "well in the field we can't be precise." (sounds like a pityful argument of "because that's what we always do" to me.)
Back to CPR though, if you have ever heard a split S2 heart tone, you know that ventilation actually degrades perfusion pressure. The drop in vascular pressure during inhilation creating pulsus paradoxis is greater than 10mmhg. So everytime you ventilate with positive pressure, you are removing any benefit compressions were giving you. Especially since the last research I have seen show both hospital and out of hospital providers baggging an average of 40 times a minute. When conciously aware of the need to slow ventilation, they slow to average of 22. If you give 1 breath in a second, 22 seconds of a minute there is no perfusion. If it takes 5 -7 compressions to build a perfusion pressure, 60/5 (assuming minimal compressions to build perfusion pressure) = 12 + 22= 34 seconds on no perfusion, while somebody is getting tired doing compressions. so basically 1/2 the time the patient is getting nothing. If you are still bagging 40 a minute, 60/5 =12 + 40= 52 seconds of nothing.
Passive oxygen should be able to maintain 11% minimum oxygenation, without the harmful effects of over oxygenation/ventilation, while perfusion is being maximized.
The theory is sound, but in EMS, old habits, especially those without evidence, die hard.
That sounds like a great idea. NRM acting as seal for O2 to reach the lungs, OPA keeping airway open, rely a little on the remaining O2 while the supplemental O2 begins to spread, and uninterrupted compressions, yeh? That's what I think the thought is behind it. 
