You are overthinking the process.

The important measure, as you stated earlier, is whether or not the individual cylinders are within an acceptable degree of difference of each other (rule of thumb, 10%). The amount of compression compared to a "manufacturers spec" usually can be ignored so long as it isn't wildly off the mark. There are so many variables to consider (ring condition, oil viscosity, coolant temp, oil temp, valve condition, carbon build-up, cylinder wall condition, etc.) that matching a spec for compression is not likely to happen in the uncontrolled conditions of your home garage. However, you can be comfortable in assuming the current conditions, whatever they are, should affect all cylinders equally in a healthy engine. So, assuming that is the case, they should all fall within a similar value band. One or more cylinders being overly different from the others indicates a problem and is what you should be looking for. Will the "manifold" of the extension affect the overall reading? Yes, but it will do so equally across all the cylinders, therefore cancelling itself out when comparisons are made.

Don't sweat it, unless you need the math practice.
--
Bob Kraushaar '94 945T, '91 940T, '88 240, '94 F-150, '68 Shelby GT-500 KR