The discussion shared below was part of many Q&A sessions Harsh Singhal conducted with Data teams at various companies and colleges.

Analytics is statistics applied with code (code = SQL/Python/R/Java/etc.)

E.g., when you identify the average or variance of an attribute in a dataset or quantify the median, you are applying statistics by writing code that interacts with your dataset.

If you learn more theory (more statistical methods), you will bring more creativity to problem-solving. If you learn more technical topics (aggregate functions, array methods in SQL, CTE, etc.), your solutions will bring performance and other improvements.

E.g.,  you may need to know a few outlier detection methods that require knowledge of quartile ranges, normal distribution, etc. to find anomalous (or weird) data points. These could be very long-running website sessions or customers that have complained lots of times.

So now that you know how to compute an anomaly will you be able to use SQL and extract the necessary results? If perhaps you know how to code in Python but you need to apply the anomaly detection method on data stored in Big Query? Also learning technical skills will introduce you to advance methods in SQL such as window functions and other aggregate functions that will solve your problem.

See Data Sketches, which combines fascinating theory (probabilistic data structures) with technology (code available to use these methods in the real world).