So, now that the results are in, it’s time to do a post mortem ! Yey !
Most important thing
First things and most important things first, I want to thank everyone who played my game, and even more important, everyone who left feedback on my game. This is more important than people may think, because every time I go to the page of my game, it lights my heart to see all the nice comments. So thank you everyone who commented, I did not dare answer the comments to avoid cluttering the page or sounding bragging/humble/clunky/etc., or just by shyness, I don’t know. But seriously, even months after the ludum dare, I like to go back to my games review to re-read the commentary, it helps me get through the hard time. So thank you, your kind comments are really important to me.
Results
Second things, my game is playable here ; https://ldjam.com/events/ludum-dare/44/immortals-game, I recommend playing it before reading this post mortem to avoid spoiler, but do as you please, I won’t force you ^^
I was really surprised by the results, I knew it would be way be way better than my previous entries based on the comments, but I was not expecting such good results : 1st in fun, and 4th overall (and in humour, and well in the other categories; in the compo category) ! I'm really happy ! :D 
By the power of friendship !
For this edition, I went to a friend’s apartment for the weekend, and even though we were working on different titles (it is the solo compo after all), this allowed us to eat together, go for walks, and just keep a kinda healthy lifestyle for the weekend, which is more important than you may think : going outside and still having social interaction is a nice way to take breaks and stay productive over the weekend. You can play his game here if you want https://ldjam.com/events/ludum-dare/44/biomass. So if you have a group of friends that want to participate in the ludum dare, why not make a compo group? That way, you can have friends, have feedback on your game, while still working solo and keeping your independence :D
Use the engine/language of your heart
For the jams, I don’t use unity or any other game frameworks. I just do it in pure dirty javascript. This may seems weird or stupid, but this is the language I am the fastest with (for game making). I do not code the best in it, but it is so fast to get something playable. I know a lot of people use unity in the LD, and this is fine, it is a really good engine, but I would recommend to people to not just use it because all the other people are using it, but to use it because you want to : I spent multiple hours trying to learn it and having the impression to fight against the engine, before I understood it was not meant for my coding style, and just accepted that I would be better with a lower level framework. So, when choosing what technologies to use for the ludum dare, choose the one you prefer, the one you know, or want to learn, and not the one everyone’s using. Even in pure javascript, you do not lose much time compared to the engines if you know it well enough. So have fun using your engine/language, and choose according to your taste, not according to what everyone else is using ! (but use unity if you like, as I said it is a really good engine)
Theme and level design
[This section has minor spoiler to finish the game, so I recommend playing it before reading this section. But again, do as you please, I won’t judge you]
This theme did not exactly inspired me at first. Making a stock market exchanging life? trading life for upgrade? At the end, I settled on the idea of dying originally to get money; I wanted to make a metroidvania since a long time, and it was the good moment to do it ! I started by making a list of the possible death idea, and a rough sketch of the level 
; as you can see, I had a few more death than implemented, but I did not had time to implement them, and I though they weren’t as fun or cool as the other. I also wanted for the level design to show the player what he must do at all time, and to try to not have him stuck; that’s why you must not unlock all death to get to the end, and a lot of interactive element/possible death are visible at the start, like the laser or the water, that the player understand will be useful later, and so when they finish exploring without item, they try to interact with those elements. In the end, the cost of the items was a bit improvised, I just knew that I wanted the player to use all the items, but to get to the exit without the bombs (so they would need to take the long road).
So in the end, I think my game had a good combination of reflexion and platforming, enough to have challenge and be fun, and easy enough to avoid being stuck, which is what you want in game jams : an hardcore game will not be enjoyable by everyone, because not everyone is an hardcore gamer. You can if you want, but not everyone will enjoy your game (and this is fine ! Not all games appeal to all people)
Polish
This time, I finished my game quite in advance : after 36 hours, I had a game playable and everything was done ! Except for polish : this is the thing I do at the end, when the gameplay is finished. And this time, I realised the enormous difference it makes when you polish the game : I added the character animation for the jumps, lots of particles effect, screen shake for the bombs, transitions for dying/restarting, better dialogs with typewritting effects, and even updated from 30 to 60 fps (yes, I realised I locked it at 30 fps at that moment, so I had to update all the physic engine to account for the doubled framerate). And holy moly, it changed a lot how fun the game was. It felt more alive, and just better ! So I recommend to anyone, polish your game ! The fun comes not only from the gameplay, but also from the polish, feedback and “juice” https://www.youtube.com/watch?v=216_5nu4aVQ https://www.youtube.com/watch?v=AJdEqssNZ-U . So to make a fun game, consider keeping a few hours to polish it and playtest it.
Conclusion
For the next compo, I will probably go with something totally different, maybe a 3d game using an engine and not in vanillajs, because I know that if I make a worse game the next time, I will only have me to blame, and if I use another tool, I will be able blame it on my learning ! So it is a win-win situation :D
This was a pretty amazing ludum dare for me, the best I made and probably the best results I will make, and thank you all for your comments, rating, annd to reading up to this point !
TL;DR : Thank you, Doing the compo with a friend (on different game) is fun, do not use unity just because everyone else is using it, think a bit about your level design, and polishing your game makes it more fun to play
; as you can see, I had a few more death than implemented, but I did not had time to implement them, and I though they weren’t as fun or cool as the other. I also wanted for the level design to show the player what he must do at all time, and to try to not have him stuck; that’s why you must not unlock all death to get to the end, and a lot of interactive element/possible death are visible at the start, like the laser or the water, that the player understand will be useful later, and so when they finish exploring without item, they try to interact with those elements. In the end, the cost of the items was a bit improvised, I just knew that I wanted the player to use all the items, but to get to the exit without the bombs (so they would need to take the long road).
So in the end, I think my game had a good combination of reflexion and platforming, enough to have challenge and be fun, and easy enough to avoid being stuck, which is what you want in game jams : an hardcore game will not be enjoyable by everyone, because not everyone is an hardcore gamer. You can if you want, but not everyone will enjoy your game (and this is fine ! Not all games appeal to all people)
None of them made the cut, but eh, they weren't that great anyway.
Its funny to say considering how few themes I voted yes on but this was a pretty good list of themes this year.