Our original crimes DataFrame was not sorted and slicing still worked as expected. Sorting the index will lead to large gains in performance. Let's see the difference with slicing done from step 8:

>>> %timeit crime.loc['2015-3-4':'2016-1-1']
39.6 ms ± 2.77 ms per loop (mean ± std. dev. of 7 runs, 10 loops each)

>>> crime_sort = crime.sort_index()
>>> %timeit crime_sort.loc['2015-3-4':'2016-1-1']
758 µs ± 42.1 µs per loop (mean ± std. dev. of 7 runs, 1000 loops each)

The sorted DataFrame provides an impressive 50 times performance improvement over the original.