Expected behavior of the map generating engine

The following is a list of expectations I had regarding the map generation protocols of the game when I started modding a year ago. I have not tested all of them, but every one I have tested either is disproven, or used to work and does not anymore:

Percentages: a percentage should refer to the % chance of something happening (i.e., probability = decimal/1.00):

• A 1.00 chance (100%) of getting planets should mean that every star system should have a planetary system; a 0.50 (50%) chance should mean that, on average, half of all stars should have planetary systems.
• A 1.00 chance of habitable planets should mean that that every planet in a habitable lane should be habitable; a 0.50 chance should mean that, on average, half of all planets in a star EDIT: habitable lane are habitable.
• A 1.00 chance of habitable planets having traits should mean that every planet has a trait; a 0.50 chance of traits should mean that, on average, half of all planets have traits.
• A 1.00 chance of habitable planets being extreme should mean that every planet with a trait is extreme; a 0.50 chance of planets being extreme should mean that, on average, half of all planets with traits are extreme.
• Thus, setting all four to 1.00 should mean every star has a system, and every system with a habitable lane has all habitable planets in that lane, and they are all extreme.

Weights: weights should refer to the proportional probability of an event relative to the whole (i.e., probability = weight/ (sum of all weights)):

• Two systems with weight 50 and one with weight 100 should show up in 1/4, 1/4, and 1/2 of all systems on average, respectively, since 50/200 = 1/4 and 100/200 = 1/2.
• Two planet traits with weight 50 and one with weight 100 should show up in 1/4, 1/4, and 1/2 of all planets with traits on average, since 50/200 = 1/4 and 100/200 = 1/2.
• Two extreme planet traits with weight 50 and one with weight 100 should show up 1/4, 1/4, and 1/2 of extreme planets on average, since 50/200 = 1/4 and 100/200 = 1/2.

Ranges with modifiers: ranges should go between the upper and lower bounds, as modified by the modifier (i.e., value = minimum + (round((minimum-maximum) * some modifier * a random number))... note that this is a linear example, I do not expect linear modifiers)):

• A system with a range of 1-5 dead worlds should have either 1, 2, 3, 4, or 5 worlds with equal chance if the modifier is neutral. A low modifier should favor 1 or 2 planets, a high modifier should favor 4 or 5.
• A system with a range of 1-5 dead worlds should always have at least one dead world (barring in-game effects).
• A system with a range of 1-5 dead worlds should be able to have 5 dead worlds, and at most 5 dead worlds (barring in-game effects).

Ranges without modifiers: ranges should go between the upper and lower bounds, and each should have an even share of the total chance of the range.

• A planet with a class distribution including a category listing of class 10 with a weight of 30 (out of a total weight of 150) should mean that that planet has a 20% chance of being class 10.
• A planet with a class distribution including a category listing class of 1-3 with a weight of 90 (out of a total weight of 300) should mean that planet has a 10% chance (1/3 of 90/300) of having class 1.
• A star system that can have a weight between 255 and 200 should not be allowed to have a weight of 50 unless the appropriate number is changed in another file.

Since these mostly do not work as expected, it is very difficult to mod this game in a consistent, much less realistic, manner. I understand this is a very low priority for the developers, whose primary goal has to be to serve the broader gaming community, but as someone who has spent a year trying to mod the game only to have these expectations disappear one by one, I am incredibly frustrated.

I would appreciate it if the game behaved more rationally, and I and others would be more motivated to continue generating content for the game if it behaved in a rational manner. Thank you for your time and attention.