Pandas can be easier for interactive poking around on smaller data sets. It feels more like Excel, whereas Polars feels more like SQL. It's no surprise that on a forum full of programmers Polars is widely preferred, but IMO it's worth at least trying both.