Orthogonal array testing - doubting the result

When doing ortohognal arrays testing, the process is like this:

Consider a function with 3 variables,each with 3 options.

3 pairs, each with 3*3 values = 27 parametric pairs. Each "function call" contains 3 pairs, i.e. in the optimal arrangement, I need only 27/3=9 test cases to cover all pair-wise combinations. I get that, orthogonal arrays provides a way how to achieve this optimal arrangement.

But when trying to apply on the following example, I do not believe the result: 75 binary parameters, i.e. 2775 pairs, each with 4 outcomes, i.e. 11100 parametric pairs to test. As a functional call contains 2775 pairs, 11100/2775=4. It seems unbelievable low but I just follow the same process.