All answers about the plus sign in Excel formulas I could find say that the sign at the front is ignored and is a Lotus 1-2-3 legacy. However, that isn’t entirely correct.

A plus sign before a range in the Excel N() function forces the function into a dynamic array formula. For example, =N(+A1:A3) will create a spill region as opposed to =N(A1:A3), which gives a single value. That makes sense, as N() is meant to convert the argument into a value.

In some other array functions, the plus sign does nothing; for example, =ABS(+A1:A3) gives the same result as =ABS(A1:A3). The bare =A1:A3 gives the same result as =+A1:A3.

However, in some array functions, the plus sign is an error; for example, =ROW(+A1:A3) is an error while =ROW(A1:A3) is fine.

Does the plus sign in =N(+A1:A3) and =ABS(+A1:A3) have the same meaning? Why it doesn’t work in =ROW(+A1:A3)?