The tie rod angles DO change. The control arm mounts stay in the same place, but, the hub moves 2" up and away from them. Which means that when everything is sitting on the ground, you have in effect lowered the control arm mounts by 2" (since the height of the wheel doesn't change). The steering arms will end up with more angle. I can take pictures of my steering arms if you don't believe me, but it does change the angle pretty significantly. Better yet though, I'll post a link up to another thread where a member actually ran my suspension geometry through a suspension program, and compared it to a stock suspended car...
see these graphs here...
It's a pretty long thread and discussion, but you can see from the graphs (produced using my suspension geometry measurements) that the bump steer was increased with the drop spindles. The first few graphs are the stock suspended car (lowered a bunch), the second set are based on my car (72b&b).
The torsion bar adjusters don't really add or subtract preload. They don't actually twist the bars. If you take a look at the lower control arms, you can see that what actually happens is the adjuster changes the angle of the lower control arm with relation to the torsion bar anchor in the LCA. This is no different than clocking the LCA's. It changes the amount of available suspension travel, but has no effect on spring rate.Torsion bars are straight springs. Unlike a progressive spring, the spring rate of the torsion bar is fixed, and constant. For example, if you have 150 lb/in torsion bars, that's always what they'll be. 150 lbs of force per inch of deflection. Now, if you're using more of that travel, then you will see higher suspension loads. But the spring rate didn't change. And of course, your handling and ride will improve greatly if you're not bottoming out your suspension all the time.