I checked the code and I see nothing that prevents picking focuses with weight 1. The weighting is skewed however and does not work like in other places that use weights like this. Each focus rolls a die from 0 up to its ai_will_do value and the largest value wins. For example if one focus with ai_will_do 2 is available and another with ai_will_do 4 the pick ratio the latter wins in 50% of the cases where it rolls above 1 and half of the remaining cases where both roll between 0 and 1. So the pick rate is not 33% vs 66% but actually 25% vs 75%. Simulating this method with all the German starting focus weights 100000 times I get ~136 times Oppose Hitler, i.e. it takes on average 735 games to get this start.
Edit: also the dice roll is not done for ai_will_do's below 1; this means 0.999 is twice as likely to be picked as 1.0
We were specifically told by the person who made the system that this is not how it works. There is no dice roll on the base value, it just takes the ai_weight of all the available focuses, adds a random value of up to 50% to it, and then picks the focus with the highest value. That means if a focus is not within 50% of the base value of the highest available focus, it will never be picked.
I decided to test this so I created a mod where I modified the generic focus tree to have a weight of 1 for the navy, industry and political focuses, a weight of 10000 for the army focus and a different value in each series of tests for the aviation focus. I then checked how many of the 57 countries using that focus tree pick the aviation focus as their first choice. Each test series consists of 10 games. All results can be seen in the attached screenshot.
In the first series I used a value of 5001, because then the aviation focus should only be taken when it gets the minimum random multiplier of 1 while the army focus gets the maximum multiplier of 0.5, assuming the official statements are correct.
The results of the first series show a chance of about 25% for the aviation focus to be chosen. This would only be consistent with the official statements if there are no steps in between 1 and 0.5, so for every focus it would be a coin flip between that focus' AI weight being unmodified or cut in half. Even if there was only one step in between the two (for example 0.75), the observed chance for aviation being picked would be too high, as it should then only be about 18%.
In the second series I used a value of 9998, which should produce exactly the same results as the first series if it is indeed only a coin flip between the multipliers of 1 and 0.5, because again only a combination of 1 for the aviation focus and 0.5 for the army focus would cause the aviation focus to be taken.
However, the results now show a chance of about 50% for the aviation focus, which is inconsistent with the results from the first series. It might be possible to explain the results if the chance or method of assigning the multipliers is somehow different depending on the difference between the focuses' AI weights, but this would seem needlessly complicated. However, the results are perfectly consistent with bitmode's diceroll theory.
I then did a third series with a value of 4000, which means if the official statements are true, the aviation focus should not be picked at all, because 4000 is less than 50% of 10000.
But, this assumption turned out to be completely wrong, as the aviation focus was still picked with a chance of about 20%, which is again consistent with the diceroll theory.
There was another developer statement in this thread claiming that the limit for focuses to be picked is actually 1/3 of the focus with the highest weight, so I decided to add a quick final test. Here, I used a value of 400 for the aviation focus, which is only 4% of the 10000 for the army focus, but in the fifth test game, Oman picked the aviation focus, so there seems to be no limit under which a focus absolutely cannot be taken, which means Germany should at least have a slight chance of opposing Hitler.
Also, I apparently have too much free time if I decide to spend it on analyzing video games like this.
. Given the random rule in 1.6, it would be important to check that.
