@madgergely , you code has a bug, if I (the same address) call makeMove twice I will win all the ether. Not ideal...

I knew it was too good to be true.

Thanks for pointing that out.

Well it seems test cases were updated since i submitted this solution, I've got this info on my solution page:

"These solutions originally passed when they were submitted, however the kata's test cases have since been updated and these solutions now no longer pass."

How do you make sure that leading zeros are added?
IE, "01:01:01" rather than "1:1:1"

This algorithm is incorrect.
It passes only because of the limited testing required on submission.
The 4, 9, 40, 90 etc that you are trying to catch with (p<c) can only be done in the first position.
Test.assertEquals(solution('CMXCIX'), 999, "Your algorithm gives an answer of 1021 and fails on this test.");

Note, this answer, although sleek, doesn't actually use memoisation.

I have created a 'fork', I suppose only time will tell if someone will do something with it.