Better gas economy - this "manual" transmission utterly confounds me

Upon reflection, you’re probably right about not being able to run a controlled experiment outside of the lab.

Here’s one cite that says lugging causes excessive wear to the crankshaft bearings. It’s right above the note in the left column. Here’s another that says lugging causes excessive fuel delivery. That means, I was also incorrect when assuming lower RPM means less fuel is being used.

Let’s think through this…mechanically speaking, I have difficulty seeing how the oil film breaks down in any car with an oil pump that puts out decent pressure. Perhaps the problem is if you’re at very low rpm and the load is high, and the oil pressure is low, then it causes a problem. So thinking through that, I could see that. So I think you could be right there. I still think that most (good) cars would have more oil pump pressure than needed. I wonder if the detonation, which is often associated with lugging, is what’s actually damaging the bearings? But if that’s true the net effect is the same, you would be correct that lugging would result in more bearing wear.

However, this is aside from the fact that 1500 rpm in a 4.6l modular Mustang GT engine is not at any point where the engine should be lugging, unless you were doing something like staying in 5th and going up a hill. If the OP’s issue is fuel economy then the engine shouldn’t ever be in a lugging situation, because if they wanted acceleration then the qualifier of fuel economy is gone and they could simply downshit.

Most cars with EFI would probably sense the detonations, and not allow them to occur, probably by limiting the fuel flow. When digging around for those two cites, I found several others that said it caused “scarring” of the cylinder walls, another reason not to lug.

I found this when looking up what you meant by a “Modular” engine (it’s one of Ford’s engine families, analogous to Hemi), I don’t know enough about cars to know if this is what we’re talking about?

I don’t know about that. My Mustang knocks plenty and has no knock sensors. IIRC knock sensors on non-turbo or non-supercharged cars are still not on all models.

Several nits to pick with various replies. (in no particular order)
Knock sensors do prevent knock by limiting fuel. They limit knock by retarding the timing. The later the spark is introduced, the lower the cylinder pressure, the less the chance of detonation. Retarding the spark really kills performance. More heat, and less power is generated. In addition on some engines if retarding the timing is not enough to kill the knock the system will inject additional fuel to kill the fire. Needless to say if your engine goes into knock enrichment you fuel mileage will suffer.
Excessive detonation can damage the pistons. I have seen pistons with a neat little hole right thought the center from detonation. It addition, the piston ring lands can be destroyed due to excessive detonation. This is called a burned piston, and the part of the piston above the rings in one part is just gone. It looks like a blow torch just burned away the piston and left the rings intact. Not a good thing, I assure you.
Lugging the engine can be very damaging. Not that it necessarily will damage the engine, but it can. With old British cars it was considered a rule that you did not put a heavy load on the engine at less than 2000 RPM. At lower RPMs the oil pressure was too low, and the bearings would get hammered. In addition the cranks could also get worn from the hammering. If you took your MG-TC and tried to climb a hill at 1500 RPM in 4th gear, you would need rod bearings in very short order. I suspect the same would hold true for a Model A or other early cars, but my experience is with old British iron. Modern engines have much better bearings, and hardened crankshafts so the risk of damage is lower IMHO.
Best fuel mileage is achieved with a combo of low RPM and large throttle opening. This is due to the lack of pumping losses when the throttle is wide open.
On a typical gasoline engine when running at steady state cruise at say 200RPM the intake manifold is in vacuum. There might be between 18"-22" of vacuum in the intake manifold. This is due to the fact that the engine wants a lot more air than can pass by the throttle since it is not wide open. So in the cylinder when the intake valve is open, and the piston is traveling down it is sucking a vacuum in the intake manifold. It takes energy to suck this vacuum. Energy requires fuel. So if you eliminate the vacuum, you will get better fuel mileage. I know of at least one engine management system that in the low mid-range (1500-2500RPM or so) the electronic throttle is held wide open, and engine speed is controlled via injection duration. This requires a more powerful processor than most engine management systems have, but I think it points to where we are going in the future.

The answer is probably 1500 in 5th, but on a modern car with variable valve timing, and various other electronic controls, 1800 in 4th might be slightly ahead of 1500 in 5th.
If I was going for max fuel mileage I would start off in 1st accelerate moderately to about 2500 RPM and then shift. On flat ground I would probably go to 3 or 4, it would depend on how the car felt as I let the clutch out. Then I would use very large amounts of throttle to get up to about 2500 RPM and shift again to the highest gear that would allow me to maintain at least 1500 RPM. Once I was in the highest gear, I would down shift if I had to use more than about 20% throttle.

On paper, an auto is not the best device to try to figure this out, way too many variables. Look to smaller piston powered aircraft. the tells ya in da book.

Best range at YY00 altitude = xxxx RPM and xx " of MAP (manifold pressure.)

**Best endurance ** at YY00 altitude = xxxx RPM (will be lower) and xx" of MAP. (will be higher)

And so you drone across the sky.