tldr: i think 4 cts and 8 ts should be possible
How the server chooses how many people are on T side and Ct side is interesting to say the least. Obviously I haven't seen the code myself but I'm pretty sure I know how I works cause I haven't seen any unexpected switches yet so. Essentially how I think it works is that the number of T's must be more than double the number of CT's and that the number of CT's gets maximized as long as the first thing stays true.
Some examples of this are:
6 people = 1 ct and 5 ts. its not 2 cts and 4 ts because then the number of ts wouldnt be more than double
7 people = 2 cts and 5 ts
9 people = 2 cts and 7 ts
12 people = 3 cts and 9 ts.
As you can see its impossible to answer the questions "If there are X T's how many CT's are there?" It can be fluid. I find that pretty dumbo head for balancing purposes.
The last example is what i find to be one of the most unfair results of how the ratio works. there are literally 3 times the ts and cts. in the first there are 5 but thats with a small population ig. and these examples are just stable configurations.
In fact before any of the stable configurations change to another state that could be more fair you need to have more ts then are actually in the stable configurations. If those ts join (or if chosen cts leave) between the end of a round and the beginning of the next then the ratio can get even more fucked up than that. Ive just recently seen 5 cts to 15 ts when the worst stable ratio for 5 cts is 13 ts (which is still bad).
basically if you want the max stable number of T's for a number of CT's double the CT number and add 3. unstable numbers can basically be whatever.
now for all of these examples we've had less than 19 ts. i suspect that the number of ts required for lr does actually affect the ratio in a way i will explain imminently but because of the rarity of 2 lr and 3 lr whenever ive seen it ive always forgotten to pay attention the the ratio and test my theory.
when i say the T num must be more than double the CT num i mean it must be at least 1 more than. I suspect that the "1" in that is actually not always 1 but really the number of Ts required for lr. obviously most of the time thats 1 but i havent been able to test this yet.
anyway i think a lot of the possible stable configurations and a huge number of unstable ones are quite unfair to the cts and there should be some change here.
How the server chooses how many people are on T side and Ct side is interesting to say the least. Obviously I haven't seen the code myself but I'm pretty sure I know how I works cause I haven't seen any unexpected switches yet so. Essentially how I think it works is that the number of T's must be more than double the number of CT's and that the number of CT's gets maximized as long as the first thing stays true.
Some examples of this are:
6 people = 1 ct and 5 ts. its not 2 cts and 4 ts because then the number of ts wouldnt be more than double
7 people = 2 cts and 5 ts
9 people = 2 cts and 7 ts
12 people = 3 cts and 9 ts.
As you can see its impossible to answer the questions "If there are X T's how many CT's are there?" It can be fluid. I find that pretty dumbo head for balancing purposes.
The last example is what i find to be one of the most unfair results of how the ratio works. there are literally 3 times the ts and cts. in the first there are 5 but thats with a small population ig. and these examples are just stable configurations.
In fact before any of the stable configurations change to another state that could be more fair you need to have more ts then are actually in the stable configurations. If those ts join (or if chosen cts leave) between the end of a round and the beginning of the next then the ratio can get even more fucked up than that. Ive just recently seen 5 cts to 15 ts when the worst stable ratio for 5 cts is 13 ts (which is still bad).
basically if you want the max stable number of T's for a number of CT's double the CT number and add 3. unstable numbers can basically be whatever.
now for all of these examples we've had less than 19 ts. i suspect that the number of ts required for lr does actually affect the ratio in a way i will explain imminently but because of the rarity of 2 lr and 3 lr whenever ive seen it ive always forgotten to pay attention the the ratio and test my theory.
when i say the T num must be more than double the CT num i mean it must be at least 1 more than. I suspect that the "1" in that is actually not always 1 but really the number of Ts required for lr. obviously most of the time thats 1 but i havent been able to test this yet.
anyway i think a lot of the possible stable configurations and a huge number of unstable ones are quite unfair to the cts and there should be some change here.