If you have a 1:1 plant that is the offspring of a CBD dominant and THC dominant plant, it has one functional THC allele and one functional CBD allele. They are codominant. The parents are Bt/Bt x Bd/Bd; the offspring is Bt/Bd.
Assume the following: (I don’t know how to do subscripts on here so I made an image)

It can be treated as a simple, single locus cross in a Punnet’s square (ignoring very rare crossover events):
The original F1 cross being Bt/Bt x Bd/Bd. All offspring are Bt/Bd:

The F2 cross of a 1:1 ratio plant is Bt/Bd x Bt/Bd. Offspring will be 1:2:1 Bt/Bt:Bt/Bd:Bd/Bd. So a 1:1 crossed to a 1:1 (or selfed) gives you a 1 in 2 chance of having intermediate progeny in the next generation, not 1 in 20. If you choose two 1:1 plants from the parent population, the rarity of those plants among the parent population doesn’t matter anymore.
