Greg,
1) It was a joke. There was a smiley in there, followed by the word "seriously." Sorry if you missed that.
2) It's a combination of the overall length of the car, but the critical dimension is the distance between the bolster and the coupler face. The longer that distance is, the more the variations in height between the two bolsters are amplified. Take two cars - one with 24" between bolsters and 6" between the bolster and the coupler (a 1:4 lever), one 18" between bolsters, the same 6" between bolster and coupler (1:3). Now, put a 1/2" difference between the two bolsters. The longer car's coupler will be offset 1/8" (1/2 ÷ 4), while the shorter car's coupler will be offset 1/6" (1/2 ÷ 3). The shorter car has the greater offset.
Now, dips in our track are seldom so clear cut, they're more often along the lines of vertical curves, in which case the longer car takes up a greater segment of the arc of the curve, which can lead to large offsets depending on the severity of the vertical curve. If you've got a train going over undulating track, where one coupler is at the bottom of a curve and the adjacent one is near the top of the next one, then you have problems, and longer cars are more susceptible to that kind of variation. Again--if you're running long cars/long trains, build your track without the vertical curves. It's not the coupler's fault.
... Dips and twists can occur on a solid foundation, wood warps, concrete heaves, free ballast shifts.
That's the joy of being outdoors. I make small adjustments to my track pretty much every other time I run due to those delightful forces of nature. It's the trade-off I make for using a more scale-sized coupler. If you don't want to do the maintenance, then you limit your choices of couplers to those that can compensate for the more uneven track. It's not that the smaller ones won't work, it's just that they won't work given the restraints you yourself are putting on the situation.
3) I'm rebutting Chuck's statement that "...
only the G-scale Kds will work."[/i] (emphasis mine) If you go back to my earlier post, I readily acknowledge that long trains bring with them their own set of issues that have to be taken into consideration. My argument is that while the G-scale couplers give you a decided advantage due to the increased height of the coupler face, they are in no uncertain terms the "only" Kadee coupler that will work.
...this is not the first time I feel you have really pushed an argument and do not have direct experience, only your theories. ...
You are correct in that I do not run long trains or standard gauge on my own railroad. That
does not mean I do not have experience doing so at exhibits or on other railroads. My experience is not limited to garden railroading as it occurs on the 300' of track on the TRR. Rather quite the opposite. My experience on the TRR is significantly shaped by the things I have learned on others' railroads over the past 33 years. The fact that my experiences run contrary to yours do not render either of our experiences invalid. They are our individual experiences. I respect yours, and I would presume you to pay me the same courtesy. I can't program a DCC decoder to save my behind, but I don't tell people it can't be done. (I generally tell them to call you.

)
I think you will find that when I don't have direct experience with something, I'm very up front about qualifying my statements to that effect. (Or, I don't respond because I haven't a clue what they're talking about.) There have also been occasions where I have stated something I believed to be true which was proven not to be, and have expressed appreciation when corrected on the matter. That is a different situation than when someone posts something contrary to my personal experience, then goes on to suggest my experience is wrong. I will defend my experiences. I welcome and respect others' experience as it adds to the collective, but I will not--under any circumstances--discount my own as invalid.
Later,
K