So let's do a simple benchmark. # without patch ./python -m timeit -s 'string=bytes(range(256));table=bytes(range(255, -1, -1));delete=b"abcdefghijklmn"' 'string.translate(table, delete)' 1000000 loops, best of 3: 0.55 usec per loop # with patch ./python -m timeit -s 'string=bytes(range(256));table=bytes(range(255, -1, -1));delete=b"abcdefghijklmn"' 'string.translate(table, delete)' 1000000 loops, best of 3: 0.557 usec per loop # keyword specified ./python -m timeit -s 'string=bytes(range(256));table=bytes(range(255, -1, -1));delete=b"abcdefghijklmn"' 'string.translate(table, delete=delete)' 1000000 loops, best of 3: 0.771 usec per loop From my observation, the difference between PyArg_ParseTupleAndKeywords and PyArg_ParseTuple when parsing positional arguments is very small. This means it won't make old code slowdown by a large percent. And when keyword argument is specified, there is a degrade. But I think this happens everywhere using PyArg_ParseTupleAndKeywords.