Same! I don’t really have any experience with selecting males or females other than hobby chucks, so take what I have to say with a grain of salt. I’m just a newbie, which is why I try to avoid picking breeding males by doing open pollination preservation runs. Females are easier - pick ones you like to smoke!
But one thing I have noticed in the preservation runs is that a large amount of the females from the F2 seeds look like their moms. Not always, maybe not even 50%, but a lot of the time the growth characteristics and “looks” of the mom pass to their daughters. I mean, this makes sense, right? It’s pretty common to have kids look like their parents, or puppies to look like their dams & sires.
I think observational breeding is mostly guesswork. Until we have good genetic testing, I don’t trust anyone who started breeding within the last few years and says every seed from their their Urinal Cake Bx3 is a dead ringer for the original mom! I just don’t think it can be done. I also don’t really think it’s worthwhile because these days, mashing together 2 polyhybrids will almost always result in at least a decent outcome by IG standards, if not a pretty good one!
My suspicion is that only a few people in the industry have enough experience and do enough testing to even have a clue how a particular cross will turn out. It takes so much experience with the mamas and the papas, so much testing, and such strict procedures to make sure there’s no cross pollination. Punnett squares and math don’t actually help beyond maybe 1 or 2 obvious traits because we don’t know what genetic traits are linked to each other, which are fully dominant and recessive and which are only partially or incompletely dominant.
You can’t really know what the male brings to the table unless you use him to pollinate many different moms that you’re already familiar with and which are genetically diverse enough that you don’t just get the same predictable hype cookies derivative outcome.
And then you gotta keep the clones around for so long (especially if doing line breeding) because thorough testing of a few crosses from each male probably takes a year+ from the time you make the seeds to do internal testing, and then probably another year+ to send out the seeds to get enough feedback to really know what the male brings to the table. And then you might have to start over if he’s too dominant, or has some undesirable traits that weren’t expressed in your environment.
There are not that many breeders doing this level of thorough testing of their males. And even fewer who’ve been doing it for the decades required to even get a good sense of how certain traits will interact with each other.
In short, I don’t believe you (not you specifically, but the general you) can look at 1 or 2 or even 3 rounds of males and females and know what they’re going to do unless they’re so genetically similar that you’re bound to get something similar in the kids. And then, why would you? All you’re doing there is derivative work, not advancing the field. There’s certainly money to be made in that approach, but it’s not really breeding, and I wouldn’t call it honest work.