Jon, do you see a market for the 7DII except for those high-end shooters looking for reach?
If they want quality they have the 3 FF cameras.
If they are looking for value they have the 70D.
First, there is more to IQ than what the sensor does. I've said this a lot before on these forums, as had Neuro. Sometimes the focus capabilities and frame rate of a camera are VASTLY more important than how good the sensor is. In that respect, there aren't three FF options...there is really only one: The 1D X. The 5D III has a decent frame rate, but if the 7D II hits the streets at 10fps, the additional 4fps over the 5D III is going to be very significant. Were talking about a 67% increase relative to the 5D III. The 6D, while it has it's advocates, and it definitely pounds out the High ISO IQ, is definitely lagging in the AF area. That is not to say the 6D AF is bad...however it's no 61pt AF system either.
If the 7D II hits the street with 10fps, a comparable APS-C optimized AF system to the 61pt system (say 41pts?), and a 20-24mp sensor, then I absolutely think it will have a market, and I think it will sell like hotcakes. You definitely cannot compare that to the 6D. The only benefit the 6D has going for it is the larger sensor...but that is one out of many factors that affect IQ. The 5D III, in circumstances where frame rate is not critically important, will probably still give the 7D II very solid competition, but in the cases where frame rate is critically important, the only real full-frame counterpart to a 7D II with such specs would be the 1D X.
So yes...it will definitely have a market. I suspect they will fly off the shelves, despite the existence of the 6D. I suspect many a 5D III owner will buy one as a backup. I know for sure that a lot of aspiring amateur bird photographers will be picking one up, especially if a new 100-400 hits the streets along with the 7D II.
The reasons the 7D was a success have not changed. Nothing has really changed. The only thing that would kill the 7D II is if it hit with a MSRP over $2500...then I think that would kill off early adopters and slow sales until the price drops below $2500 at least (I think a price around $2000 is most likely.)
If Canon wants to replace the 1D (and IMO that's the only reason for 7DII to exist), Canon will try to replicate the IQ as close as possible. And that will mean lower MPs.
Note that Canon was pretty conservative with increasing megapixels on their 1D line.
They must have noticed people who want reach are wiling to sacrifice resolution for light sensitivity.
Mind you, the difference between APS-H and C isn't as much as with FF, so innovation in sensor light-sensitivity might well allow the newer APS-Cs (maybe not 7D though, if it still has the 70D sensor) to trounce the 1D line.
Why would you think the 7D II is a replacement for the 1D? The two lines are designed for completely different groups of buyers. The 7D II is not, has never been, and will never be a replacement to the 1D X. Your completely ignoring the price difference here. The price of the 7D II is likely to be less than 1/3rd that of the 1D X. It doesn't really matter how good the 1D X is, or how good it's potential successor is...price is the real segregator here. The 7D II will sell because it will fall within the realm of generally or eminently affordable to the majority of middle class buyers. At $6800, the 1D X does not fall into that class, not even close.
Regarding the megapixels of the 1D X, Canon didn't get away with anything, nor were they conservative with anything. It's so funny how short peoples memories are. I remember, very very clearly, the loudest and most vocal outcry from Canon customers before the releases of the 1D X, D800, 5D III, and D4. The loudest demand, by a very LARGE margin, was: "Fewer megapixels! Better high ISO!" That was what people wanted, was VERY MUCH what people wanted from the 1D IV/1Ds III successors. Canon didn't skimp, get away with, nor was conservative with the 1D X. They delivered EXACTLY what their customers literally demanded. For those who can afford it, it seems to be a raging success as well. It's converted more than a few pro Nikon shooters to the Canon camp, and has made raving fans of existing Canon 1D/1Ds users.
There is no chance the 7D II is ever, ever, ever, ever, EVER going to "trounce" anything with a larger sensor. Just not gonna happen. Again, outside of the one caveat I mentioned, the IQ you get from a sensor really has nothing to do with the pixels. Improvements in pixel technology are primarily going to reduce read noise, which is already a very small factor in the deep shadows, and secondarily improve sensitivity. The quantum efficiency, Q.E., of the 7D is already at 41%. To double the noise performance, Canon would have to achieve 82% Q.E. I know of sensors that can achieve 82% Q.E. at around 650-600nm, but all of them require significant cooling to do so...usually a Delta-T of -50°C or more. That kind of cooling requires a lot of power (for example, most high Q.E. Astro CCD cameras use two- or three-stage peltier cooling...a peltier, or TEC is a thermoelectric cooling device that uses a P-N transistor matrix to create an electronic heat pump...they draw significant amounts of energy.)
Now, assuming Canon did somehow manage to improve the Q.E. of their sensors to 82% at average operating temperatures (pretty much not possible, dark current is going to be very high at operating temperatures of 60°F and above) but just for the sake of discussion, let's assume they do. At 82%, they can double their noise performance (reduce noise by half). That still doesn't cut it. The 1D X has a total sensor area that is 2.6x greater than the 7D (and any potential 7D II), so you would at least need to increase the 7D II's Q.E. by 2.6x instead of just 2x. Well, that isn't actually possible. To double Q.E. again beyond 82%, you would need 164%...but you can't have better than 100% Q.E. (quantum efficiency is the ratio of conversion of photons to released electron charge in a photodiode...you cannot convert more than 100% of the incoming photons.) You can't even get that .6x better than 82%, because that would require another 50% Q.E....or 131%.
All that matters in this particular area is total sensor area. So long as Canon does not make the 7D line use at least a FF sensor, it will never be able to "trounce" the 1D X in terms of equivalent IQ. In terms of reach, all that matters is how big the subject is reproduced on the sensor...in which case smaller pixels mean more detail, but total light gathered for your subject is roughly the same regardless of sensor.