Heh, our solutions were both counting sort (mine was only so rigorously microoptimised it may not have been transparent), and it pulls 10049 ms in comparison to your 11000, so it wasn't a problem with my code, but rather your tests being too difficult.

I've changed the tests a little to be lighter (at least the completion time dropped for me from 12 to 11 sec).

Great, hopefully that helps alleviate the issues in your second comment. Thanks for responding quickly and helpfully. When something like just time of day can change whether or not a valid solution passes it really isn't great after all, at least a 50% success rate with a decent solution I'd hope, and no successes at all was what had made me particularly concerned, twenty seven consecutive failures is really quite a lot. Trying again it worked first shot though with a comfortable margin at about 11.3 seconds, so the little leeway to allow for things like that is nice.

Definitely agree it still could warrant a kyu raising though, although I have no idea how that would even happen, I've mostly just been solving katas around here, not helping make them.

particularly for something that is supposed to be a 6 kyu

It's not me who chose 6 kyu, altough I wouldn't call this a 5 kyu as well.

across 27 attempts and six solutions I had not a single attempt succeed

This depends on the server load, received inputs, and Python version you're using. When the estimated completion time is 12 sec, even good solutions are not guaranteed to work all the time.

the kata seems to be too tight and difficult

I've changed the tests a little to be lighter (at least the completion time dropped for me from 12 to 11 sec).