You score 80 on a history test and 80 on a chemistry test. Same number, but the history class averaged 72 with a standard deviation of 4, and the chemistry class averaged 74 with a standard deviation of 12. In history you are 8 points above average, two typical gaps: z=480−72=2. In chemistry you are 6 points above, half a typical gap: z=1280−74=0.5. Your history result is the standout, because the z-score uses each group's own spread as the ruler.
The formula does two jobs in order. Subtracting μ moves zero to the mean, so the sign shows which side you are on; a negative z is not an error, it just means below average. Dividing by σ changes the units: centimetres, seconds or points all become 'standard deviations'. That is why z-scores have no units, and why a whole data set converted to z-scores has mean 0 and standard deviation 1.
A z-score is not a percentile. z=2 does not mean 2% or 'top 2%'; it means two standard deviations above the mean. Only when the data are roughly normal can you convert, and then z=2 sits near the 97.7th percentile. For a strongly skewed data set the same z can match a quite different percentile, so check the shape before translating.
Think of it like this. Converting to z-scores is like converting Fahrenheit to Celsius: 86°F and 30°C are the same heat measured from a different zero with a different-sized degree. A z-score moves the zero to the mean and makes one 'degree' equal one standard deviation, so values from any group fit on one thermometer. The conversion keeps order and spacing, not the data's shape.
Core idea. Measuring a value's distance from the mean in standard deviations puts values from any group on one shared scale, with the sign showing which side of the mean the value falls on.