1. If the crouch bonus to aimed shots for sniper weapons was reduced, it was only because before it was very, very big before. It's still larger for sniper weapons than any other weapon except the RPG. I think I just wanted to make sure they were still effective long-range weapons when standing without being too accurate when crouched.
There's actually no crouch bonus (crouch=1) for the snap shot, up from the same 0.5 as the aimed shot—although I can certainly understand the desire to tone down the snap shot, I feel like from both a design and a physical perspective (more on the latter in a bit) it makes sense to maintain a simple "crouch --> improved accuracy", even if that factor ends up being 0.9.
2. Snap shot accuracy for sniper rifles was reduced because they are big, heavy, unweildy weapons that can not be easily aimed quickly. Think of this not as a measure of the weapon's accuracy, but of the capability of a soldier to accurately deploy the weapon with the given fire mode.
I'm dubious about this line of reasoning for two reasons. First, I'd generally argue that it's too divorced from questions of balance to be good design (and if we're really insisting on realism, a better starting place would be quintupling all weapon ranges and most map sizes), and second, it seems circular—it's a big, heavy, unwieldy weapon because the developers collectively say it's a big, heavy, unwieldy weapon, not because of straightforward consequences of the setting and general technology level.
Actually, I think a third issue is that that "big, heavy, unwieldy" bit seems to already be the turf staked out by TU use—so by increasing the TU use by 3 /and/ scaling the accuracy way back, it feels like we're now talking more along the lines of "anti-materiel rifle" rather than "sniper rifle" in the "big, heavy, unwieldy" department.
3. Machine guns were made less accurate when crouched because the strong recoil makes the position less stable than a standing position. Try going to a kneeling position (our new models will be in a kneeling position) with your right leg down. Your back leg (right) is not as capable of pushing back against heavy recoil (driving into your shoulder) as it is if you are standing. You must rely more on your back muscles. Rapid fire of a heavy machine gun requires strength to keep the fire on target, and this is why standing is a more stable and accurate position.
I understand how that view could be arrived at, but I don't think it holds up on a physical level—machine guns really don't push backwards enough to require the kind of bracing you're thinking of, the issue is controlling the motion of the weapon's barrel. According to the HEAD UFOPedia, the machine gun uses a 4.7x33mm tungsten-cored steel round (which is, incidentally, /tiny/), which now that I think about it I'm too lazy to do the precise math on right now so let's take a substitute. The .50 BMG round, though not usually sporting tungsten, is substantially larger at 127x99mm; we'll take mass of the the heaviest variant (52g), the muzzle velocity of the fastest variant (928 m/s), and the rate of fire of the M2 Browning (635 rnd/min). That works out to a force of about ~510.71N during firing. Assuming the soldier masses 100Kg (including gear) and taking Wikipedia's listed approximation for the coefficients of static friction for rubber on dry concrete, static friction will top out at ~980N—it isn't necessary to brace at all in the manner you're describing. Wet concrete tops out at ~294N, but again, this is full-out uninterrupted fire with generous assumptions about the force exerted.
So if we accept that the issue is barrel control, I'd argue that a kneeling position is clearly more stable—you can brace the lead arm against the lead leg, lots of postural sway is eliminated, you've got more contact with the ground, and the position permits leaning forward to get more of the long axis of the weapon pressed against the torso. At the very least, it strains credulity that the stance should be less accurate.
~J
