Reddit
What makes a Ludum Dare game best overall?
I was talking with my friend earlier today and I was wondering what is the most important thing to focus on a game (after functionality): is it graphics? Innovation? Audio? Something else entirely?
So I wrote a quick script to show the correlations between all of our statistics and overall. The script is here.
The correlations are (lower is more correlated):
{'Humor': 179, 'Innovation': 66, 'Community': 158, 'Theme': 107, 'Graphics': 73, 'Fun': 23, 'Audio': 251}
Fun is obvious. If a game isn’t fun, then it’s not going to be good overall. But fun is also pretty vague and hard to define – it doesn’t give much more information than ‘overall’.
Innovation and graphics come after – these are things that are more understandable. So, conclusions: work really hard to come up with clever ideas, and spend a lot of time on the graphics.
Pretty sad to see audio come in dead last, since that’s something I worked hard on 
Interesting, especially because “spend a lot of time on the graphics” goes against most of the advice I’ve seen for participating in a competition like this.
The implicit category that’s missing is ‘functionality’. Without functionality, graphics are meaningless
True that.
I think it’s not so much “spend lots of time on graphics” as it is that [functional] games with good graphics imply that the game’s creator knows what s/he is doing- through use of libraries and frameworks that make graphics easier, for example, or for knowing shortcuts to make basic graphics look awesome.
Alternatively, if people all follow the common advice – use extremely basic graphics first, focus on making the game playable, then worry about good graphics last – then that would mean all games with good graphics also have everything else (because everything else should get done before graphics). It’s probably not that simple either though!
Rather than just sum the squares of the differences, could you calculate the Spearman’s Rank coefficients?
First find the mean score in each category, and then use the formula in the link:
http://en.wikipedia.org/wiki/Spearman%27s_rank_correlation_coefficient
uh, sorry, I mean Pearson coefficients.
A = 0
B=0
C=0
For each i
{
A += (X[i] – meanX) * (Y[i]-meanY)
B +=(X[i]-meanX)^2
C +=(Y[i]-meanY)^2
}
Output A / (B*C)^0.5
http://en.wikipedia.org/wiki/Pearson_correlation_coefficient
Low correlation for audio might be caused by the fact that only a limited number of games have anything resembling audio, so audio won’t get rated.
Seeing the low correlation in Audio and Humor, but a lot of games don’t actually have any audio or humor in them. I think it would be interesting to first throw out all data on audio and humor where the game got more than half N/A votes in those categories. That usually means there wasn’t any.
Yeah, I was thinking this too. I may release a better version sometime, or maybe you can
I think that most important factor in LD games are gameplay: innovative gameplay (aka fun&overral) are most important thing, followed by theme
TA-DA!
http://i.imm.io/8Dbo.png
Interestingly all pairs of variables have positive correlation, not just with “Overall”.
Now this is even cooler than mine. Well done!
One question, what is the scale? The line is sloped downward which seems to imply negative correlation, but that just can’t be true!
Top left is (1,1). Bottom right is (4,4)
It’s interesting to wonder if these correlations apply to life generally.
eg, When we choose our experiences, more than anything we want our experiences to be Fun. Fun is an emotion based experience.
Our second highest priority is Graphics, ie we want our experiences to be beautiful to our our eyes. Note that vision is our dominant sense. Graphics is a sense based experience.
Our third highest priority is Innovation, ie we want our experiences to propel us into new territory to satiate a thirst for new concepts, theories, mechanisms and ideas. Innovation is experienced intellectually.
ie, Emotional experiences > Sense Experiences > Intellectual Experiences
I think Jung would agree :p