Bear in mind that fuel consumption doesn't matter in this case, only air and water consumption. I am also assuming that the throttle is closed completely when the throttle pedal is not pressed (I'm not 100% sure on a fly-by-wire car, though I think it's a safe bet).
In neutral the engine is spinning at it's lowest possible speed (~800rpm-ish) so it's consuming the smallest possible amount of air, which means that it will ingest the smallest possible amount of water because the flow rate through the intake is minimized.
If one is using compression braking the engine is spinning at some speed above idle, say 3000rpm, even with the throttle closed it will consume more air than if it was spinning at only 800rpm, and my intuition says that the engine will ingest more water as it consumes more air.
The fuel efficiency argument really depends on the car, I have noticed in my MS3 that the AFRs spike high during engine braking (over 30:1 afr, maybe there's a fuel cut?) so it probably consumes less fuel than if it was in neutral (~14.5 afr).
(2cents)
Edit: I have noticed MAP pressures of around 19-22kPa (absolute) during both engine braking and idling, so the intake is under vacuum in both situations.
As always, thanks for taking the time to answer, James. I had a fundamental misunderstanding when it comes to the AFR. I was under the impression that at idle, the throttle was slightly open to allow some air to mix with the fuel in order to keep the engine moving along. This is not the case and under both engine braking and idle, the intake is under vacuum. Which makes sense because the throttle is 100% closed (what I thought happened at idle is actually a big problem!). As described below, this is probably handled by a valve that is separate from the throttle blade.
In regards to what you noticed with the AFR on engine braking vs idle, I found an interesting site where someone explains how a Bosch throttle works. It sounds like this is relevant to your interests:
The other function of the idle switch on later models is fuel cut-off during coasting (over-run condition). This fuel cut-off occurs until the engine RPM fall to a set value at which point fuel is restored for idle condition.
http://www.firstfives.org/faq/throttle/throttle.html
Some other places I've been reading seem to correlate that thought of fuel cut until closer to idle speeds when the throttle is closed.
Forgive my ignorance, but if the throttle is closed during both engine braking and idle, should it not matter which is done through water because air should not be coming through the intake? Then again, if at idle fuel is being squirted into the cylinder, air is required for the mixture. When the injectors are shut off during engine braking, there should be no air coming in with the closed throttle.
I did some further reading on idle air control valves and it seems like this is something that would open up to keep the car idling, where the ECU operates a valve that allows enough air in to idle. During engine braking, I don't imagine that the ECU would open this valve because there is no fuel to mix with. Then again, I'm a noob at this stuff. Here is a relevant link to the idle air control valves:
http://www.se-r.net/engine/about_iacv.html
Now the follow up to the idle air control valve is how is something like that situated on the MS3 (or any other car)...does it take a small bit of dirty air, bypassing the intake, or is the air for this valve coming through the intake? If the former, then the conversation is moot because any air being taken up isn't coming through the filter close to the ground.
I should get one of those Dashawk things...
This with one person in the car. Should be interesting going to work in the morning with my wife in the passenger seat.
The upper grill looks a lot better bro, very nice work. I think you need some sort of lip. What do you have in mind?
