If you use the entirety of your account to support a position, you are flirting with disaster. You have no reserve and cannot kick in the afterburners to get out of Dodge if the need arises. You should have some reserve buying power to get out of trouble if the need arises.
With gold shooting for the moon, i am not sure you can do well with credit spreads consistently. A few highly speculative bets maybe, but other than that you are probably better off sticking with the indices. Debit spreads would be a much better way to play gold in my opinion. But i wouldnt touch the put side.
Your point is valid but I think it was just a contrived example to highlight the particular margin implications he was discussing. You are always allowed to close any existing position regardless of reserve buying power left. Reg-T ensures this. MoMoney.
Bottom line: Look at your statement. Margin is not re-computed everytime you add and remove a position -- it's computed daily based on settlement prices for that day. There is no difference between a one step or two step process in the rollover in terms of margin. You don't "loose" margin --it's already lost.
I have tried that, and it just doesn't work well at all. I called for support and the tech I spoke with agreed, that it doesn't work with things like Remote Desktop.
If you rolled for a credit (i.e. roll to a future month) then buying power would not be an issue, and you could roll to a new spread with a position size of 10. I don't see how you can do that with a two-step roll...
LOL. My mistake, I assumed you you were rolling up/down. However, the same principle applies when rolling to a different month for a credit: If your account allows you to roll in one step, it will allow you to roll in two steps with the exact same size [EDIT: Don't take my word for it! - try it out, I'd be glad to be proven wrong] MoMoney.
LOL, since I'm bored, I'll explain the small flaw in logic: As the front month spread is now worth $5, let's say the back month spread is $5.50 Take the $6000 in your account after locking in the loss on the front month, add to that $5,500 credit you get from selling the back month spread and you have enough margin to satisfy 10x 10-pt spreads. This is why I prefer to close a losing spread rather than rolling to a further out month and end up with a position you might not otherwise have put on. Either way you are locking in the full loss on your front month spread. Any perceived credit for the roll is just the difference between the front and back month spreads but I'm sure that was obvious. MoMoney.