I would like to thank you all for this awesome result! :smile:
Plug Me is my best Ludum Dare entry so far (here is the chart of my past Ludum Dare).
I will for sure expand the game and release it on Steam, so if you want to know when it is ready just follow me on my social (Facebook or
Twitter) or just follow me on my main site havana24.net! :stuckouttongue:
Even if I think that the main prize in ludum dare is to have a brand new game in your hand I think that a better judging system will improve the overall experience for everyone.
That's why I thought a new rating system that I want to share with you and maybe start a conversation in the comment section.
The main problem of Ludum Dare rating system
From what I see in the comment section of the results post the main problems are two: 30% of games doesn't get the 20 ratings and some complaint about people get to the top leaderboard with "just" 20 ratings.
To be more fair, in order to enter the leaderbord, LD should increase the minimum number of rating to like 30 or so. But as we can see LOTS of games don’t hit the minimum 20 ratings required, just because the developers don’t know they should leave feedback/ratings to other games in order to get other developers see their game.
So, to me, instead of increasing the minimum number of ratings, they should impose a minimum number of games that you need to review in order to enter the final leaderboard.
My idea to improve the Ludum Dare ratings
First part of the system.
* You release the game in time for the competition.
2. You get redirected to a page where the system explain that you need to rate 20 games in order to enter the leaderboard (with like a fancy bar at the top that loads every time you rated a game).
3. You press a button and the system automatically redirects you to the next game you need to review.
4. Once you leave all the ratings and a feedback for that game the system allow you to press the “NEXT GAME” button in order to get all your 20 reviews done.
* If you can’t test a game because you don’t have the required system (like you need Windows but you got a Mac) you can press the “SKIP BUTTON” and select why you didn't play the game (maybe the system can make a report to the developer so he know why people are skipping his game).
* In order to prevent people skipping all the “bad” games you can implement a maximum number of skip.
Obviously you can also play and rate all the entries, but to be eligiable to the leaderboard you need to review games that the system gives you.
In this way the system could choose the right games that needs to be rated and you can spread ratings on every entry.
But wich games the system should give priority?
Here comes my second part of the system.
* *The first two weeks of ratings * the system will chose games with low ratings in order to get them to the right number of ratings, letting them to enter the leaderboards.
* * The last week of ratings * the system will always choose games with low ratings BUT ALSO games that have a high score in overall category (and have less than 30 ratings) in order to reach a more higher and reliable data (like 30 ratings or a bit more).
Basically here the system is trying to avoid that games with exactly 20 ratings will get to the top without having enough data, spreading also ratings over all the top games in order to push them with near the same number of ratings.
I'm sure that there is a point where if you get "X" ratings (I put 30, but it could be higher) or 50 ratings, or 100 ratings, or 300 the final result doesn't change.
We just need to find this number.
Other things that could be implemented
* Pass from a 5 stars based ratings to a 10 stars based ratings. *
To me 5 stars are too restrictive, sometimes you need to give the same amount of stars (especially in the overall rating) between two games even if one of them is better than the other, just because there is no half-way.
We don't want to pass from 5 to 10? Just introduce fractions (3.5, 4.5 etc).
* Comments should be anonymous *
With this system lots of people tend to be more accomodating and kind because they are afraid of a "revenge" (speaking about ratings).
In fact lots of people complaint about their very low final rankings because they see many polite and good comments in their games that don't reflect the actual ranking.
Basically commentators post "Good game" and put 1/2 star ratings in some categories without helping the developer, without leaving a comment to really improve the game.
If we put anonymous comment we will get more true feedback that refects the actual rating, but we will lose the "back to back" ratings behavior.
* Introduce a forced comment if you put 1 star rating or 5 star rating * (or 1/3 and 9/10 if we use the 10 star rating)
This will provide more valuable feedbacks to the developer in case of a low rating and prevent a little bit the blind "5/5" without leaving a comment.
* Let the "Theme" category influece the overall final result *
To me in a jam the Theme should be the main focus. Because without following that theme you can "easily" make a great game (even before the jam start) and put yourself in a great spot in final results.
So maybe make a median between Theme and Overall? Or a median where the higher weight is based on the Overall but also the Theme count?
That's my 2 cent.
What do you think about?
_ -havana24 _
https://ldjam.com/events/ludum-dare/39/low-speed-internet-space-robot-adventure