To make a long story short, I might be getting a 75w MV postline ballast soon , and if I do, I want to set up a decorative post top light in the front yard. Since the lamp is going to be visible in the fixture I am considering, I would like to use an 80w A23 lamp (available on 1000bulbs) instead of an ED17 lamp. Are the lamps interchangeable or are they different voltages and won't work with each other?

I would imagine 75/80W being that its only a 5 watt difference would not cause much harm, although i may be wrong there, slightly overdriven and slightly underdriven lamps... not sure how the life would be affected ...

Arc voltage and current specs may be different, leading to even opposite results from what would be expected..

Just as example, a 70W HPS lamp is underdriven, not overdriven, on 80W Mercury ballast. And the ballast is overloaded (though not very severely in this case)

An 80w mercury lamp is 115v, 0.8A.
A 75w mercury lamp is 130v, 0.64A

So the differences are more than just a 5w difference.

The ANSI ballast codes are:
Mercury Vapor
40w    – H45
50w    – H46
75/80w  – H43
100w    – H38
100/125w – H44

But since Power = Volts x Amps, then

115v x 0.8A = 92w for the 80w MV and,
130v x 0.64A = 83w for the 75w MV.

Power = Volts x Amps x Power Factor

For most HID and Fluorescent lamps the power factor of the lamp itself (regardless of the ballast used) is around 0.85

With a capacitor in series with the lamp, wouldn't the power factor be .9 or better?

This is the power factor of the lamp itself as would be measured across the lamp, not including the gear

The power factor we normally talk about (that can be high or low with different setups) is of the entire system, as would be measured across the line power input to the luminaire

With a capacitor in series with the lamp, wouldn't the power factor be .9 or better?

Ironically with capacitor in series, so a lead type ballast, the lamp power factor will be even lower.

The power factor is defined as a ratio between the real net powertransfer vs the product of rms voltage and current.
Mathematically it says, how much the voltage and current correlate to each other. And so unity power factor means they bave the same frequency and phase of all harmonic components. In other word the same frequency, phase and shape of the waveform.
With just linear coils, capacitors and resistors on a sinewave supply the shape of all signals remains sinewave on the same frequency as the supply, so the only possible cause for lower power factor is the phase shift. Because most real things in the AC power distribution are linear inductances and resistances, that is all what electricians think of, when speaking about power factor.

But discharges are a nonlinear devices, such they virtually convert any current feed shape into roughly rectangular voltage. That means the lamps generate higher harmonics in their voltage.
With just inductive impedances as ballasts, the higher harmonics see higher reactance, so practically do not generate any harmonics in the current at all. With resistive and mainly capacitive ballast some of the higher harmonics are not filtered out, so they form an extra current component in the feed, but in a different phase than generated with the lamp.

Because the higher harmonics are not present in the current, or are in a wrong phase, they do contribute to the rms values of the voltage and current, so contribute to the Vrm×Arms product but they either do not form any power. Hence the power is smaller.

With plain inductive ballast, so a sinewave current and a rectangular voltage the power factor is exactly 2×sqrt(2)/pi=0.900316...
The ballast with a series capacitor tend to feed the lamp with rather spiky current,where the spikes add up quite significant,y on the rms (the nature of the squaring), but not that much with the power (just linear dependence,whe the voltage is constant inthat section of the square). As a consequence, the power factor of about 0.7 or 0.8 is not that uncommon with e.g. CWA (adds its own spikes due to the saturation based magnetic regulation).