As of now I've received emails for two MAGIC games, so it is now officially a "contest". I haven't played more than a couple of seconds of each one to make sure that they work so I can review them all on an equally fresh basis, but I'm looking forward to playing them all tomorrow.

For me tomorrow is the hard part of the contest; where I have to rank all the games to see who is the winner. While I mostly think everyone is a winner for entering the contest (only mostly because I have an urge to kick my own arse for writing something so trite), I do have to do the judge thing for this to be a contest. I had a scoring scheme already worked out with Prozak to enable us to come to a consensus on which game we thought was best, but now since I'm the only judge things have got a fair bit simpler. I'll still use the same scheme since I put a fair bit of effort in devising it, but it means the final scores will be arrived at much quicker without the need for consultation between the judges.

The only issue I'm yet to make up my mind on is how to return the results. My judging scheme involves giving each game a numeric score to help aid communication between judges as to their opinion on the game (back then I wasn't sure how many judges I'd have). So when I release the results, do I just list all the entries in rank order, first to last, or do I also release their final score?

The reason I'm undecided on this is that the scores themselves have absolutely no purpose other than as a metric for ranking games. And unfortunately the laughable excuse for game critics in most publications (on-line and off-line) have re-enforced the "high school English paper" mindset of grading, where games get at least 50% just for running on the target platform. I find it annoying that a film critic would give a good, well-crafted fun film three stars out of five, whereas a game critic gives the game equivalent something like 92%. So there's a danger that people will misconstrue the purpose of the scoring and start making assumptions that just aren't true; for example, I see nothing wrong with giving the average game in a contest a score like 2/10 if it does the job of ranking games in order.

So if the scores are essentially just a ranking mechanism, why even consider releasing them? Well, as game programmers with INT stats of 16 and above I guess we all just like allocating mostly meaningless numbers to things. And I suspect people might be interested in exactly how their game went.

There's also the option of releasing exactly how I derived the scores, but I've been advised by other contest organisers that that might lead to more problems than it's worth. There's always a subjective element with grading something creative, so a lot of the scoring categories have very vague fuzzy boundaries where a judge can use their discretion. I've had enough experience arguing with students on projects on why they should have got 3.5 marks instead of 3 in category D to know the potential headaches with going with a complete score breakdown.

In any case, if I do decide to release scores you can be sure that they won't be out of something as conventional as 10 or 100 where there's an automatic assumption of a "passing grade". I might as well scale them all to be out of 1337.

Any opinion on whether releasing the final scores is a good or bad idea? As I'm teetering between the the two right now your arguments could sway my opinion.