To supplement PGmath's answer:
It is worth noting that the math is calculated a bit differently than you might think. It has to do with limitations of storing and representing floating point numbers (reals) in a computer (in binary format).
For the same reasons you cannot represent
1/3 in decimal exactly (it is 0.333...) you cannot represent exactly (for example)
1/10 in binary (but it is easy in decimal as 0.1).
The decimal floating-point numbers you enter are only approximated by the binary floating-point numbers actually stored in the machine.
This can result in some surprises. You can see what the number stored in machine actually is like this in a Python Console:
from decimal import Decimal
Decimal(2.675) = 2.67499999999999982236431605997495353221893310546875
So when rounding this number to 2 decimal places it will result in
2.67 instead of
2.68 you expect.
When building large nets of Math Nodes such "errors" can lead to a butterfly effect causing big differences in the end on the output.