Don't trust Excel. Got it.

Thanks again for your help!

"Harlan Grove" wrote:

Rothman wrote...
There must be something about it not liking having to subtract irrational
decimals (i.e. =cell-cell)

I've found another formula to get the same numbers that doesn't trip up the
mode function (just from the nature of my data, there's a simple average that
could be taken instead).

Still, this is frustrating.
....

Formulas like =X99-Y100 are a very good example of formulas that may
appear to produce the same results but are just slightly off after
floating point rounding error. If you believe two numbers should be the
same at, say, 3 decimal places, then you need to round them to 3
decimal places. Figuring that Excel (or any other software that uses
finite precision floating point math) will do the right thing and that
you can trust what it displays is very fertile ground for floating
point bugs.