Hi there
As you rightly say, the results are simply dictated by the votes given by the other participants.
I had a protracted discussion about it here but -long story short: results, ranking and score are a bit of a misnomer.
The whole "voting" phase is just a bit of extra fun where participants get to hang out and recommend entries they like. In turn, such entries will appear higher in the list, as a way to suggest them to casual visitors. (Which in turn tweaks the main page ranking, which is sorted by popularity/views).
So to answer your question, there was really no method or evaluation or parameters, it's just a casual recommendation system not much more intellectual than a Facebook thumbs-up.
As discussed in the other thread, most people seem to enjoy it for the little bit of fun it is (we have 20+ entries in first place as ex-equo with two/three votes each after all).
But both you and one Russian team seem very concerned about it, so we're re-evaluating for the future.
I believe the contest needs a way to reward originality and creativity. This seemed a fun way to achieve it, but I realize the whole idea of a ranking (no matter how loose) is uncomfortable for some.
Probably I will replace it by recommendations made directly by myself as a jury of one, picking the 10-odd entries that look more striking or collected more votes (regardless of their language).
I don't like the lack of transparency that entails, but it seems worth exploring instead of giving no recommendations at all.
Thanks again for joining the contest and have a lovely day 😄