You could yes.

Yep we decide by looking at the individual strike price probabilities.

Usually anything inside 20 days to expiration is far game but of course it depends on the situation.

We would always manage that side before expiration to avoid assignment since most assignment happens the last week of expiration.

Correct which is why we don't go inverted when we move up spreads.

Thanks.

Stop losses are crap - here's the backtesting data on it: optionalpha.com/stop-loss-option-strategies-21775.html

Rolling the losing side increases risk. You close the spread with a guaranteed loss, then move it up/down but what if the market continues to move against you? Do you close and roll again - locking in loss after loss after loss. Why not just leave it along and take whatever loss you have and manage the unchallenged side. Make sense?

There are several ways you could roll the losing side. One way, which you pointed out, is to buy back the losing spread and then resell further away. How large a loss you take depends on when you exit the losing spread and how far out in both time and price you resell the new credit spread. Yes, if the market continues in that same direction unabated, you could be locking in loss after loss. But if you are selling a put credit spread, the volalitility will also pick up when the market drops, allowing the next spread you sell to be sold farther away and/or for more money. The spreads can be sold very far away in the indices each time you roll. The tradeoff is that you increase probability of a small loss or smaller gain after rolling the losing side at the cost of not defining your risk. The second method of rolling the losing side actually reduces risk. Instead of rolling the entire spread, roll only the short strike while keeping the long strike. Each roll either gets closer to the long strike reducing your risk or even past your long strike. You could even end up with a situation where you roll past your long strike so that each subsequent roll of the short strike makes you more and more money if the market continues in one direction.

Believe me I understand the rationale behind what you say here but in reality (since we've backtested more than 21million different trading strategies, adjustment techniques, etc.) is that rolling the losing side is not as effective as rolling the un-challenged side closer.