I rather marveled at the foldout probe on the Datron Neo and have been working at designing one up but recently it occurred to me as a temporary measure I could use a high quality Picatinny (Rifle Scope) rail and quick release scope mount ring mounted to my Spindle head bracket that I could quickly and easily install and remove with repeatable accuracy without always putting it in and out of the spindle. After a bit of trial and error I have a really solid repeatable level of accuracy, ~.001, that I can pop in and out very quickly. I stress again a good quality rail that is precision machined and matching rings allows you to change scopes and sites with high degrees of accuracy to make a minimal amount of scope/sight re-cal.
I am curious about what is the best way to then program and store the offset between the center of the probe and the spindle center in Mach3. There is a ton of info about the offsets for a number of things and I'm a little confused at to which offset and tool I should assign this to as a matter of best practice.

You could clock a hole or boss with a clock in the spindle and zero out X&Y and take a reading of the machine position figures.

Then probe the bore/boss with the probe and then you will have another set of figures the difference between them can be stored in 2 DRO's on screen for you to adjust the position by or hard coded into your probing routines at the end of the calculations.

Or you could pick a completely different method.

I'm thinking a button with a G10 L20 Xx Yx  or simply a G0 X**y** then Zero and I suppose it could be G5x...

As I said there are many ways to do it.  Its the one that works for you that is best.

I figured many have probably done something similar and perhaps there was a 'tried and trued'  method perhaps...good intel is always a great first step...thanks!

I just want to follow up after running a large number of cycles while I sort out my offsets how amazingly accurate this has turned out to be...I have a nice flat profile on the bottom of the spindle I have tucked the 1" circular mount I machined tight up under and have filed it down carefully to get an almost uninterrupted stream of exact repeats after removing, reattaching and running the center-finding macro over and over with never more than a .001 deviation.