Whew! Tom that is a beeeeeeeeeeeg list of Q's. I'll try and be as systematic as possible working through this.
I'm curious, if I take pollen from a certain type and pollinate another type, have you seen that the resultant plants tend to be more like the seed parent, or are they random like basic genetics would dictate?
Very good Q! My experience thus far has shown that seed vs pollen parent definitely has an influence on progeny. I would strongly recommend starting with line breeding to create a pedigree line that has stable traits. In essence, creating a homozygous line. Whilst this is a tedious process it does generate a breeding outcome that is the foundation for further projects. I don't know of any professional breeder that does not use this modus. This germplasm is basic building block of their breeding programmes.
If you apply this to something like Petunia or Pelargonium it all makes sense - all the present outputs still rely on those initial breeding lines which are still backcrossed onto more recent additions to provide new recombinations. It all sounds like a heck of a process (which it is!) but I cannot stress enough the value of going this route as opposed to random outcrosses with heterozygous parents.
The downside is the space needed for evaluation of progeny. This will however be substantially less than the random method. Because your parents are line bred they are uniform - therefore there will be a limited number of recombinations available as opposed to heterozygous parents with a broad range of possibility (because the parents are a random combination of dominant and recessive traits). You can therefore take a random sample of your seed because the chances are very high that you will have all possible combinations present. From these selections one can further refine the breeding lines to achieve the desired product - either through line breeding or further recombination.
I want to cross a chocolate with a red. I can't be sure when I hand pollinate a flower whether it was self pollinated or if the cross worked, so I figured if I got chocolate peppers from a red pepper plant, or the other way around, I could be more sure of the cross. If the cross is truly random, it wouldn't matter which way I cross them, and which pods I collect seeds from.
Remove the anthers (before they dehisce) from the seed parent - there is no way it can self then. Also if you want to 100% sure of parentage I would suggest using one plant as a seed parent only - removes the need for tagging flowers. Use a young plant and only pollinate the lower-most flowers and remove all the rest until you berries have ripened. Better to be sytematic and be 100% sure.
Another aspect, chocolates seem to be hotter than reds, but they seem to have a more limited use. I would think that a world record red would end up being more popular. Since reapers and the new one were both made from nagas and peppers like that, I would assume that if I bred two peppers that are hotter, and much hotter than reapers there is a good chance at something hotter than any of the HP numbered hybrids.
There is empirical evidence that the progeny cannot be hotter than the parents and any subsequent line breeding attempts will see the progeny aligning to one of the parents' capsicanoid content. Unless one has hitherto unknown knowledge of gene recombination in Capsicum, the chances of breeding a megahot from a superhot and a non-superhot are unlikely.
It therefore makes sense to start with those that are documented to have high capsicanoid content - with a high range where the lowermost readings would be a million plus SHU and the higher random reading spike above 1.6 mil. I would look at the Alphanerdz Douglah, Moruga Scorpion thingy, CARDI Scorpion (original germplasm available from the NuMex Chile Institute) and BrainStrain proper. My reason for recommending these and not things like Naga Morich and Bhut Jolokia is that in outcrosses where these were used (Primo, Butlah etc) the heat came from the other parent, they contribute a better flavour/aroma though.
It is a bit more complex than that though - best discussed below.
So, if chocolates are hotter than reds of the same type, why haven't we had any world record holding chocolate peppers yet?
There are two parts to this answer.
Part of the answer lies in the test itself! Your capsicanoid receptors are very different to the test that is used. Browns are known to have higher dihydrocapsaicin content than reds. Dihydrocapsaicin is responsible for the deep burning of the throat, palate and back of the tongue. It is hell fire! It is really the chemical that makes supers what they are - who can ever forgot Neil @ THSC's videos with the now iconic "Back of ma throat". However. Capsaicin itself has by far the highest volume. Capsaicin is responsible for the dry scratchy irritating burn - typical in frutescens and annuum. So in essence they test for overall amount of capsicanoids in the berry. Which does not necessarily equate to what we perceive as heat. In theory a berry could be mega high in capsaicin and would hurt but not deliver the same burn as one with high dihydrocapsaicin and medium capsaicin.
