Ok one last time with numbers
Let’s say the mean is 100 [ it is]
Let’s say idiots are 60
Genius is 140
Let say they all breed only with each other let’s say the offspring results are
Idiots average = 80
Genius = 120
They both regress to the mean but they also have more intelligent or less intelligent children on average
For your broader point to be true the average for both would need to be the actual average for all [100] it’s not [ it may even need to be low average seeing as you said more likely to have idiots]so whilst it regresses to the mean [everything does] your broader point about more likely to have idiots is false.
The fun thing is that if you’re intelligent, you’re more likely to have mediocre idiots as children….regression to the mean and all that.
You are more likely to have mediocre bright kids but not idiots. Idiots are more likely to have idiots. Given enough time /generations [ I assume more than 10 generations possibly more not sure ] and enough sampling you will get everything returning to the average/mean assuming the sample size is large enough
Amusingly [ and I wish i had read it before posting] that is what your own “supporting piece/ link” shows as well
FWIW i have not seen research with real world figures [ would be interested] with bright parent and offspring over generations to see what happens but “selective breeding” for intelligence would clearly work and I dont think you start with idiots as parents
Chuckles some more [ deffo leaving this time* as tom is belligerent if not correct – to be fair you are sort of correct here]
* i pinky promise