Python: new test frameworks to be used.

Python: Random tests are vulnerable to input modification.

The test code needs serious refactoring:

1. Currently user code is called multiple times for the same test. The same user result should be held and let different test functions inspect them.
2. Each specific test should not generate 1 test assertion for every check, that's causing too much output. Just check the same thing across all the relevant sections of user result and fail by first failure/pass by all success. 1 test assertion per test function
3. Then, the random tests should be uncommented and added back. Currently there are random tests in the code by they're not run ;-)

Needs sample tests

the kata is fairly challenging, although when a user is undergoing through 848 tests should get more information.
The fact of having half-season instead of full-season is making it complex to tackle and has been the thing I was struggling the most with (example: 10 teams, 5 teams play 5 home matches while the other 5 teams only 4).
I think that due to complexity and poor -visible- testing (no testing in user window), very few other people will likely complete this kata; I, for one, solved it because I've been doing round-robin schedule in the past and know how the algorithm works.