[Walt Fricke]

VanValkenburg gives the formula for sway bars. As a practical matter, the bar diameter is what it is, the constant determining the bar's deformation is fixed, and the bar's length is fixed. The only variable at the track is the arm length. And that is a "length squared" term. So you can't just use teeter totter lever calculations to determine the effect of changing the arm length.

As a practical matter, all you need to know is that the changes are non-linear, because you have only three holes, and you will want to experiment with all three of them to see which works best for the rest of your setup. In bars with slider adjustments you just keep moving one direction or the other to optimize.