Serhiy, here are the better performance measurements:


$ ./python.exe -m pyperf timeit --duplicate=1000 -s z=0 z+0  # list-add
.....................
Mean +- std dev: 17.3 ns +- 0.3 ns

$ ./python.exe -m pyperf timeit --duplicate=1000 -s z=0 z+0  # master
.....................
Mean +- std dev: 17.2 ns +- 0.3 ns

$ ./python.exe -m pyperf timeit --duplicate=1000 -s l=[0,1,2,3] [0,1,2,3]+l  # New branch
.....................
Mean +- std dev: 92.6 ns +- 1.7 ns

$ ./python.exe -m pyperf timeit --duplicate=1000 -s l=[0,1,2,3] l+[0,1,2,3]  # Old branch
.....................
Mean +- std dev: 99.8 ns +- 1.4 ns


Honestly, I'll defer to you here. If you feel like pursuing this is a total waste of time, I'll just close the issue/PR. Otherwise I can explore how often the branch is hit in production code, as you suggested.