When the simple average fails
If two groups have different sizes, you cannot just average their means. Each mean must be weighted by its group size.
Worked logic
A class of 30 averages 80; another class of 20 averages 70. The combined average is:
Note it is not — it leans toward the larger group.
The lever intuition
The weighted average always sits closer to the mean of the bigger group. Use this to sanity-check: with more high-scorers, the average must be above the midpoint.
Same idea everywhere
Weighted averages also model average price across different quantities, average speed across different times, and grade point averages across different credit hours.