Understood, thank you. I'm not sure weather "selling premium" is an industry-wide term which implicitly refers to just the extrinsic value but I get very well what you mean. For me it's entirely irrelevant if I buy/sell an ITM or OTM option since it's only the extrinsic value that counts, all the rest being hedged away through a position of 1 or 0 in the underlier.
Short put and higher strike long call is a bull risk reversal. Long same strike call and short put is equivalent to long shares. The "risk" in the name denotes that you cannot arb the position by shorting shares against the combo; it converts the position into a synthetic bear vertical (short put at x; long call at y; short stock).
I'll try to address each of your questions. I'll try to address each of your questions. Do I try to avoid assignment? With covered calls and cash secured puts I don't really care, but my preference is to roll the short option. Mostly what I think about is where that leaves me on Monday. Is a longer duration easier to roll? Not sure what you're asking. A longer duration option that's deep in or out of the money might be have less of a bid/ask spread than one that's near expiration and is deep in or out of the money. At any rate, it's not something I think about. Will I roll both legs of a calendar? I'm only talking about covered calls and puts. I'm not super fond of trading calendars (or butterflies). So, you'd be better of talking to someone who's really good at it. Generally, I put them on and when they're up with any kind of decent profit, I get out. I never stay in them until near expiration, and I don't roll the short option. But, again you should find someone who knows. Would the roll result in a credit? With covered puts and calls I don't care if it's a debit or a credit. With diagonals I'd consider rolling for a credit before I'd roll the whole trade. It would depend a little bit on the bid/ask spread of the long contract, and a lot on where the long contract is relative to the current price. Fear of getting assigned and not generating premium? I think you're asking about being assigned on Friday and missing time premium decay during the weekend. I think it's a legit consideration. I've had a couple of market makers tell me that they often take the weekend out of their pricing models on Friday mornings. So, it's possible that the weekend's time premium decay may already be reflected in Friday's prices. I can't say that that's true or if it really has any impact. Again, it's not something I worry a lot about. The market exposure (or lack of it) on Monday is the bigger issue I think.
So with a part of my “income generation” portfolio, I sell a 60DTE SPY put every week. The idea is to diversify by distance and durations. If volatility spikes, I will try to extend both distance and duration because I am still getting paid “enough”. The problem comes with environment we find ourselves in now. Volatility is low, we are moving up and in a dangerous location - extended into all time highs. So my fear is that we have small crash (no V recovery) and I get assigned on all my puts at relatively high prices. At this point I will be stuck - can’t generate sufficient premium by selling calls and can’t take advantage of the pull back. Edit: This is something I jus started doing, so not much real world experience to speak of
The really arcane answer, but the options are almost always better from the loss side than stock from a tax standpoint. Time erosion in an option is volatility and part carry. Buy a $200 stock today and sell it a year from now for $200 - have you lost any money? Yet you can't expense the stock carry. Options are dividend adjusted so no real benefit to either side for dividends. Price an option with vol. set to zero - if the model you use won't do a zero vol. play with it. Either set a very very low volatility of a zero rate and you can see the monetized carry in the option.
I think your approach is pretty sound, and I share your concerns. I find markets like this more stressful than a lot of bear markets.