I’ve been meaning to reply to this for awhile, and while SavageJack’s response was spot on, here are a couple more thoughts, @Deadend7786:
You are right that the deck construction rules mess heavily with the traditional odds of poker hands, but it does so equally across the board. You can make it much more likely for those full house/4oaK’s, as well as straights/flushes/straight flushes. You are also right that straights and flushes alone cannot keep up with full houses and 4oaKs, and that’s exactly how it should be since they are a lot easier to make, even with the skewed odds - in fact, you could make your entire deck one suit, and get 100% chance of a flush. You have to either have massive, consistent hand rank bumping, or make those decks shoot for straight flushes, which is extremely difficult to beat when you are backing it up with cheatin’ resolutions.
I never even attempted to make a straight flush deck for a very long time, viewing them as simply inferior. I learned, however, that it is difficult for most decks to consistently get legal full houses or 4oaKs. That’s why combined with cheatin’ punishment, even a simple flush can often do very well against those types of decks. Even with cheatin’ punishment, if you let their bullets get out of control they may be able to consistently get hand rank 7 & 8 without cheating, so that’s another consideration as well. That’s why you have to tailor your shootout actions to your goal: if it’s to get a straight flush, you need as many studs and bullet bonuses as possible. If you’re just going for a flush backed up with lots of cheatin’ punishment, you want bullet/stud reductions for the opponent so they can’t make their legal rank 7 and 8’s (my PTPT deck is an example of this, combined with hand rank bumping).
Anyway, long story short - there are lots of considerations, and I believe the various draw structures are extremely well balanced and nuanced in ways that I’m sure I still don’t fully understand yet. I did write up an article for the Gazette that goes into more depth on this topic, and is followed by many good comments from the folks in this community: