I use a single point of contact, versus a cradle, to suspend the bow handle. Mine is a lag bolt wrapped in leather so I have a surface of about 1/2" to rest upon. I find I can balance the bow at about 1" above center or thereabout and adjust the position of my draw hook to fine tune the balance. My draw hook winds up somewhere below the final nock point, between an inch to zero I'd estimate.
The primary method for determining how well the bow is balanced, in terms of the balance of stress between the upper and lower limb, and even the bend within a particular limb, is the set taken from shooting by hand. The earlier you start shooting the bow by hand, versus working it on the tree, the better matched your finished bow will be to your shooting style. I generally start shooting at least 4" before final draw lenght. And I try not to work the bow much on the tillering tree, or shooting for that matter, until I'm satisfied the load is well balanced and each entire limb takes an equal amount of set from work. I also prefer lots of shorter workouts, with enough rest too fully recover the set, so that I don't permanently set the bow with some inherent flaw early in the tillering process.
All this being said, I'm not sure how it relates to hand shock to be honest. I'd rank outer limb mass and low stretch string above tiller "timing" (more or less positive tiller) for it's relative influence on hand shock.
Perfect tiller, in as much as it can be achieved, has other, more compelling justifications anyway, predominantly premature work fatigue and the overall design requirements. More specifically, a really short bow should bend generally more on the inner limb, and longer visa-versa, imho. I find full arc of circle tiller not necessarily as desirable in as many cases as not. In fact one could make a compelling argument full arc of circle might be more likely to contribute to hand shock than other forms, based upon the idea (demonstrated in high speed film) that it's the mid limb area which tends to distort or "bulge" the most at the end of the power stroke. Full arc of circle would tend to exaggerate this, I would think, by virtue of the mid limb being necessarily weaker relative to the other two forms.