Will be good to mention in description, that random movement change(?) has to be distributed 50\50 (+-5%) between directions.

This kata would've been a lot better, and a bit tougher, if draws also counted games that had empty spaces but were unwinnable.

The possibility of a jagged matrix (i. e. lines of variable length) defeat the concept of "instructions scattered over a ‹two-dimensional plane›". It would only make sense to receive the input as a 2D matrix (i. e. a list of lists of fixed length).

not as simple as i thought

Thank you for your answer.
Yes, it was my fault.
I need to be attentive to details.

It's expected your answer to equal expected answer, so the expected value in the error message is 1 not -1. -1 is the value your function returned. Not a kata issue.

[[1,1,1],[0,2,2],[0,0,0]]: expected -1 to equal 1.

The test expects -1, although the correct answer is 1.

Might just be me but I didn't think it was very clear that the input could have variable length rows & columns, and that the expected behaviour of "wrapping around" would be from e.g. index 8,10 to 7,6 when moving up if 7,6 is the furthest out index in that row. Could perhaps be made more explicit that this is the case.

lack of random testing

so once i removed the prints everything worked fine:P

would it not also be a cat's game(no win con) before all tiles has been filled?
eg: [[1, 2, 2], [2, 1, 1], [0, 0, 2]]

Java traduction

Nice kata!

approved