Its not that it takes a huge selection to breed good weed. It takes a large selection to fix the traits and make a stable strain. If you want to find the same flavors and highs, and make truly predictable results it requires large populations. Not to say amazing weed can't come from a few good beans.
Lets clarify something.
You can find a great hybrid plant by dumb luck, then clone it indefinitely. That's what most of the "clone only" lines are. . .individual excellent plants that some grower recognized as great, then perpetuated as clones. The more plants you grow out, the more likely you are to find exceptionally good ones; that's just the "bell curve" in action. If you want to find that "one in a million" plant, the more plants you comb through, the more likely you are to find it.
But in my opinion there is a significant difference between identifying excellent individual plants and breeding. Breeding implies doing directed crosses, in an attempt to isolate and stabilize traits. Doing selections from large numbers can certainly help, and finding that one in a million "magic" pheno can be useful, but neither of these two things may be necessary. It will really depend on what you're starting with and what you're trying to accomplish with the breeding project.
For example, lets say you wanted to create an autoflowering version of a popular strain. In reality, all you're really doing is introducing just one single trait to an already existing and hopefully already stable line. In a case like this, a significant proportion of the offspring of your crosses would be expected to possess this new trait (say 1/4 of the F2s), and therefore you shouldn't need to do selections from thousands or even hundreds of individual plants to isolate it. . .only a dozen per generation will probably be enough. It will still make many generations to stabilize the line, but the actual selection part here should be fairly straightforward.
Kona Gold
I'm talking about breeders(sensi, serious, and others) that claim to grow out thousands of seeds. They feel this technique is far superior to breeders that use a smaller selection process. But with soooo much to select from, i feel it's possible to lose site of your goal, where on smaller selections you are able to maintain your original ideas and goals, potential leading to more distict strains and better consistancy.
The exact opposite is true. If you know what you're doing and actually have set a real goal (rather than the "lets cross these things and see what happens" approach), the more plants you have to select from, the more likely you'll be to identify excellent plants with all the traits you want, and the easier (not harder) it will be to either reach your goal, or to reach a fair conclusion that you won't be able to. While no doubt its physically (and possibly emotionally) harder to do the actual selection with larger numbers, selection from smaller pools typically entails more compromises. The bigger more famous breeders don't use large numbers because they don't know what they're doing or because they're "lazy". To the contrary, its a LOT more work, cost, and risk to run the big numbers projects. They do it because the larger the numbers they're selecting from, the less dumb luck plays a role in the outcome of the project, and the more likely they are to have a "breakthrough" that makes the project worthwhile.
On the actual numbers, its not just a claim. Doing a selection from several hundred plants during a proper breeding project is pretty typical. Remember, roughly half your plants will be male, and most of those get culled early. Depending on the nature of the project, you're also potentially maintaining a clone of EVERY plant, so that you have a "backup" and you can immediately start work on the next generation after you've made your selection. If you wanted to do an F2 selection from 50 female plants you'd expect to have to grow around 100 plants to sexual maturity.
I think most serious breeders would tell you they'd actually want more plants to do the early selections from, but 100 plants is the "magic" number that invokes Federal jurisdiction, and its typically a line most American growers/breeders respect.
Why so many? Lets say you wanted to isolate one recessive Mendellian trait. You'd only expect to see one plant like that out of four grown from heterozygous F1 parents. Now lets say you wanted to isolate THREE such traits (or one, that was controlled by three different recessive genes). . .you'd only expect to see one plant with all three traits out of 64 plants. That's already a already a larger number than the size of your selection pool. Even if you had 64 plants, you still might not get that pheno you're after, just by dumb luck (EG its possible to flip a coin twenty times, but NOT get three heads in a row. . .it happens). To ensure a high likelihood that you'd get it, you'd want many MORE than just 64 (maybe 2-3x as many). Without going through more math, in a nutshell, the larger the number of plants you've got, the more likely you'll be to find the rarer combinations you're looking for.
Anyway, if you want to do the same intensity of selection for F2 and F3, you're already potentially up to 200 total plants grown, more if you're cloning them. As you get further up the generations (F4. F5, etc) the plants should become more inbred, the variability less, and therefore the size of the pools you need to do your selection from can be significantly smaller. But add it up and you're still potentially looking at several hundred plants/project.
This, by the way, is why true breeders CAN'T suddenly bring 20 new "strains" to market, not real strains, anyway. It just takes so many plants grown, and so much time invested in growing and selecting for even ONE good project, that no one person can run more than a small number of these at once, meaning no more than a few new strains per breeder per year. If you look at the bigger "name" breeders, most of these guys weren't doing everything with their own two hands. . .they'd be working with a team to maintain the plants, cut clones, etc. Otherwise, things become unmanageable.
I just crossed White Widow and Berry Bomb. Good times
