Hello all. This one is aimed at those of you that have a working knowledge of this glorious plant’s genetics; of course, any voice is welcome that’s rooted in, and/or supportive of, a desire to be a better steward of the sticky.
I understand the purpose behind pheno hunting as many seeds of a breeding event as possible toq increase the probability of finding a gem in that particular crossing. Given that any and all crosses are still, at the most basic level, individuals of the same species, can we apply this principle of numbers to multiple packs of multiple breeders; assuming they are of similar quality and integrity? Say a guy has 100 beans spread out amongst breeders like Karma, Cap, Lucky Dog, Sunken Tressure. Alien, Jaws, etc. And a another guy has 100 beans of a single crossing. Is the probability of finding a winner the same? Or is it limited to each pack individually? (Winner defined as a plant with unicorn properties of cultivation and finished flower.)
It all depends on what the grower is considering a “winner”. I would say that the guy who have 100 seeds of the same cross have a better chance to find a winner, because he can look through more plants of the same and find the best expression that cross can have. The other guy will never be able to see deeper into the genetics of each cross and he won’t know what a “winner” of the different genetics actually look like. He will have a high chance to find one combo that got a better expression, but he won’t know if it was “the ultimate expression” of that cross without growing out a large nr of seeds.
Pz 
Probably best going back to the basics of Mendels pea experiments and observation in order to fully understand.
Winner is defined and its an all around qualitative description not so focused on any particular terp profile or other characteristic delineated to a level that a punet square would help. It’s asking for prediction of probability to be solved without defining the winner “beyond best plant out of this hundred lot of beans” and is something unicornish in anyone’s reasonable opinion essentially.
You’d be far more likely to run into a ‘unicorn’ when hunting through one varietal. That’s my opinion, at least. With one example you’re using 6-7 sets of likely unrelated genes and the other you’re working inside 1 genome. I guess it’d depend on how far they’ve taken that one cross? If it’s already an inbred line, then chances are probably higher with the other set, but near impossible to predict. It’s definitely a solvable problem if you were to get to the point where you can identify the cuts. While there’s likely to be more genetic diversity in the group with multiple varieties, sometimes it’s good to know your boundaries so you know what you run into. With one varietal you’re exoring much more of the genome than you are with the others and that gives you much higher chance of finding the anomalies. Fun thinking. Thank you for this.
variance will likely be higher across the many breeder/breed sampling vs. the single cross.
your chance of a unicorn may be higher in the multi breed attempt.
that is what I do, however, that doesn’t leave me with a solid feeling when i select one as there could be some better in that line.
your chance at multiple unicorns will be higher in the single breed run.
oddly shaped distributions could reverse this.
Let’s just look at a “winner” as a physical expression. Like freakshow, this may be over simplified but for the sake of imagery.
Freakshow x Freakshow = 100% offspring with freakshow leaves
F1 Freakshow x non-Freakshow = 0% offspring with freakshow leaves
F2 Freakshow x non-Freakshow = 25% offspring with freakshow leaves
If 10 F2 seeds are used you have 2.5% chance of a winner
If 100 F2 seeds are used you would have a 25% chance of a winner
beyond that. there are exceptions to everything. I have read reports of indivdule clones showing differnt phenotypes when brought differnt growing conditions. Iike up the mountain range instead of down at sea level.
This may be way off base, but I suspect the odds of finding a “winner” are, on the whole, the same.
That said, you’re more likely to find a winner for the one cultivar used in the single-cultivar test run, and less likely to find a winner for any of the individual cultivars in the multi-cultivar test run.
Seems to me there would be a lot more subjectivity involved in assessing winners out of a multi-cultivar test run (there’s subjectivity in either scenario, but with the multi-cultivar test run you’re inevitably judging “the best” out of unalike things).
Nice stimulating question 
It’s the most complicated angle to explain in the equation in fact.
Better to start with a humble example to treat the vicious trap it is : a lambda grower with a 10pack of a strain. The specimens are launched and an exceptional female is found, then a motherplant is born, then eventually a clone that is shared.
It’s not hunting. When you hunt, you know your prey. You know what it eat, how it poop, its period of reproduction … you have datas.
It’s just a quantitative screening (yes, even with 10) and the choice of the word lottery is quite neutral in this context and type of methodology.
Imagine the same grower, in the same context but with three hangars and thousands of hundred of seeds … it still apply, it still a quantitative screening.
Now imagine this grower again, enough stubborn to launch again the subsequent line he created (let’s say a F2). He don’t only learn by repetition, but also by the consequences of his choice (bad or good) on the initial pairing (P1 generation).
So even if it’s just F2 at this stage, he learned to hunt the progeny of a specific couple. It’s when breeding really begin, with mapping.
To don’t repeat what have already said Lonely, i extend to another consideration : the big factor that is the knowledge of the grower on the genetic that he’s hunting, and the “strain science” behind this.
Let’s say without making free marketing for all, that the grower only chosen among all these lines and breeders a specific source. Let’s say the skunk#1.
In this context, maybe the goal is to directly start with parallel lines already made for him, to reinforce the skunk#1 at long term. It ask a good knowledge of the outcrosses used in all skunk#1 hybrids lines, but it still an option.
And if the grower choose to only do an “all in” in a single source, he have maybe a constraint of accuracy that can’t be reached with a spreaded effort.
Still not breeding until you mapped the line ^^
The more worked a line is, the higher the average of quality is. This comes at the cost of fewer extreme phenos aka gems.
I favor running more of one line myself… that said a lotto is a lotto and the maths are pretty consistant hah
Thank you all for indulging in my very cerebral Chem4 stone 
ya’ll are good sports, and well informed. For some reason I thought it was a good idea to sort through my vault and replace desiccant packs after a few solid bong rips and got stuck on that when seeing all those packs of killer genetics. I spent a good three months reading everything I could find about breeders before I bought a single pack, now a few years later I need a bigger boat or an intervention.