Games are scored on the following aspects:
The Overall score is the weighted average of all aspects: 5 × (Theme + Enjoyment + Aesthetics + Innovation + Bonus) / 4.5, clamped to the range 1 – 5.
Having multiple aspects encourages voters to think about games from different angles and write more specific feedback. This helps jammers learn what worked and what didn't. A single score would be simpler, but it wouldn't tell you why a game scored the way it did and the outcomes would be more subjective.
Many categories were considered, but these were chosen to cover the key qualities of a game while balancing different team strengths — so that a "programmers team" won't necessarily win over an "artists team".
"Theme" discourages premeditated games and encourages creativity. Themes make jams more interesting, because you get to see how different people interpret the same idea.
"Enjoyment" captures how the game feels to play. It could have been called "Fun", but games can be stressful or sad and still be enjoyable. It tries to capture two ideas: do you want to play it again, and how much did it make you feel?
"Aesthetics" rewards games that are visually cohesive, have good writing, suitable sound design, and feel like they belong together. It could have been called "Graphics" or "Sound", but those labels make people focus on looking "pretty". A grungy sketch style can be just as effective as polished 3D art.
"Innovation" rewards trying something new. Novel games are more memorable, even if the mechanics aren't fully balanced — which is common when experimenting under time pressure.
"Bonus" captures everything else. Sometimes there's a quality without a name — how all the pieces fit together, or some excellence where "5" just isn't enough. Because it's highly subjective, it's weighted at 0.5 compared to the other categories.
Scores are calculated as a simple mean (average) of all votes. We considered using a median or excluding outliers, but either approach would make the scoring harder to understand or more likely to produce ties.
A score becomes meaningful after 10 votes — at that point a single vote can shift the result by at most ~5%. After 30 votes, which is the standard threshold for a large sample, the result is quite stable. In practice, most games receive significantly more votes than that.