What angle should I set the VTA on my VPI turntable?

lewm,
Rest the stylus on the platter at 0 grams. Lower the pivot point. I guarantee the pressure will increase. Raise the pivot point and it will loose contact with the platter. VTF is set with the scale at record level because setting above the record height (tail down) will give a false heavy reading up to .25 gm. on some scales.

cleeds,

If you raise the "tail" (pivot) you can only increase VTA.

I know. Show me where I said differently.

Increasing VTA by raising the "tail" (pivot) will also increase VTF.

Wrong. Again, it changes the balance and the VTF lightens.
As an example, visualize the arm at level neutral balance above the platter. Raise the platter until it contacts the stylus. Still 0 grams pressure. Lower the tail. Now you have pressure. You’ve LOWERED the tail, INCREASING the pressure. There’s your physics.

lewm,
You must include a pivot in your analogies or they don't apply.  Imagine raising the pivot point until the stylus lifts from the platter.  The arm will balance at less than vertical.  That's because the leverage at the cartridge decreased as the cartridge neared the gravity line and equilibrium was reached (balance).  The VTF is then zero.  It must have decreased.

bifwynne,
I see you upgraded to the VPI stainless arm with VTA adjust.  My experience with the original stainless unipivot is this: When VTA is changed, adjusting the tail up decreases VTF.  Adjusting it down increases VTF.  A lot.  If VTA isn't close, the adjuster won't make much difference in sound.  The fix was procedural.  I began using a VTA/VTF gridded block to level the cartridge and closed in on the VTA from there, readjusting VTA then VTF and repeating.  With the Lyra I imagine the situation is compounded by its' sensitivity to VTF.  You may have arrived back at the sweet spot of coil alignment but not SRA.  When in the ballpark, the range of the adjuster will take you from murky to shrill and finding the VTA sweet spot is easy.

cleeds,
Wrong. On a unipivot the balance changes and I am correct.  You're comparing VTF and VTA.  Two different parameters.  
There is a typo though. There is no VTA/VTF block. It’s a VTA/Azimuth block. My bad.

I've done the work, taken the measurements.  Nuh-uh is not a rebuttal.  Show me.

cleeds,
The sofa analogy ignores the physics of a pivot point. Teeter-totter would be better.

cleeds, slaw, lewm,
I’ll attempt to clarify. Here’s my model:
Put a dot in the center of a paper. That’s the pivot. Draw a vertical line through it. Thats’ gravity. Draw an offset horizontal line through it. That’s the tonearm. Add counterweight and cartridge. Of course we can balance it at 90 degrees. We can also balance it at 45 degrees. Q - How is that possible? What keeps it from just swinging vertical? A - Leverage. Leverage varies with the distance of the cartridge to the gravity line. Maximum leverage at the cartridge is with the tonearm at 90 degrees to the gravity line. The tonearm will usually be less than 90 degrees, it’s in the design. Lower the tail and the tonearm will move further towards horizontal, increasing the leverage at the cartridge, increasing VTF. Because the counterweight is on a much shorter lever the change in distance to the gravity line is negligible.
P.S. cleeds,
I’m playing nice. You should try it sometime.

Funny, my experience is the opposite.  I wonder if the answer flip flops depending on which side of parallel the tonearm is.  Above, below or horizontal.  I know what my arm as it's set up does.  Physics isn't my thing and there could be (and most likely are) other explanations I've not considered.  I would love an answer from someone who actually knows.  No one here does, including me.  It's fun and didn't get too sideways.  Cheers!