No problem - happy to take another crack at it. I know that I sometimes need to hear things different ways before I actually hear them
Let's approach it with an example using a comparison between the $100 spread size vs. $20 spread size. If you're sizing your overall positions using $$ then a $100k position will have 10 contracts ($100 spread size, $10k/contract) or 50 contracts ($20 spread size, $2k/contract). The $100 spread size will have a higher credit per contract - that's easy to believe as the insurance contract will be much cheaper when it's $100 away vs $20 away. But the higher credit per contract isn't 5x higher, so there is a higher total credit with the smaller spread size (All of this based on using the same total position size).
If one where taking positions based on contract count, then 10 of the big spreads will be $100k at risk vs. $20k at risk with the $20 spread size. Sizing on contract count doesn't make much sense to me. In practice I'm changing my spread sizes to balance out risk, contract count, and total at risk, with a strong bias to big spreads (say $100+ in today's IV and share price environment).
When the shares go down these two chunks of $100k at risk behave differently. With the $20 spread you go from a $100k value to $0k value over a share price move of $20 (not counting the credit - you keep the credit as an offset to the $100k loss). Good news - the $20 spread size yields a bigger credit, so the max loss is smaller! But not THAT much smaller
The $100 spread size goes from $100k to $0k over a $100 move in the share price - the max loss happens more slowly. This increases the time to react, increases the range for taking a partial loss, and makes for a bigger share price range in which management techniques are effective (effective = constrained to a net credit).
We've discussed elsewhere that spreads can be rolled for a net credit down to the midpoint of the two sides. Conceptually this is easy - at the midpoint there is an equal sized change in time value on a roll leaving a $0 credit (the bid/ask slippage will turn it into a small net debit). And in practice the net credit rolls get worse and worse as you approach the midpoint.
I figured this out for myself by going into the option chain and using the share price at that moment, I pretended that I had a variety of spreads that I wanted to roll. I had to do the roll math for myself - there wasn't something convenient to do it for me, but it wasn't onerous. I also tried different spread sizes. Examples - with shares at $710 I might roll a 620/720 ($10 ITM) and a 700/800 ($90 ITM) as well as a 660/760 ($50 ITM). I'd also try a 600/700 to see what a close OTM/ATM roll looks like (these are a lot more fun to contemplate
).
Part of what informs my spread sizes today is I had a $20 spread open that saw both legs go ITM. I took the loss early before both were ITM and got out with only 40-70% losses (2 or 3 iterations, the entry and exit prices which didn't vary a lot which was why the range of actual % loss was so large). The shift from max gain / OTM, to 1/2 loss, to full loss happened in such a tight share price range that it only needed a few trading hours to happen (which is also what actually happened). If (when) that's happening on expiration day, being away from your desk and monitoring the share price can turn a max gain position into a max loss position while you're eating lunch. Ugh.
Choosing when to roll, at least the way I do spreads, is no different from choosing when to roll short puts or calls. But that's another reason I use the larger spreads - if I go $30 ITM (which I got to when the shares went down briefly to $660 last week or two) then I might not do anything (which is what I did, more or less). It depends on my conviction and what I want to defend against. If that was a 670/690 instead of the 590/690 I had, then it would have been a lot harder to sit on my hands (or roll straight out for time). In fact the straight out roll would have required a net debit, so I'd have need to roll to a 675/695 or 680/700 (worse strike) to get a net credit.
On the smaller spreads I don't have any personal rules for how and when to roll, partly because I had that 1 very bad experience and I'm not seeing any reason to create that possibility again
. No experience to use for finding those rules.
Another point I like to emphasize - I view these vertical spreads as a form of leverage we have available. The example I've been using - a $700 cash secured put will reserve $70k cash (let's assume an IRA - I know that margin accounts will have directionally similar, yet different analyses) for the contract. If I do $100 sized spreads, then I use $10k each, or I can do 7 of these instead of 1 csp. That's 7x leverage! It isn't 7x credit as I'm buying that leverage by way of the insurance part of the spread. But 7 of these might get me 4x the credit relative to the cash secured put.
A specific example - I like the $700 share price as a strong resistance. Let's sell the 600/700 put spread for 9/3 vs the 700 strike put. The 700 strike put is $8.15 (midpoint of b/a) and the 600 strike put is 1.80. In this instance I can use $70k of reserved cash to collect $815 (csp) or 7x (8.15 - 1.80) = $635 * 7 = $4445 (math might be off - approximately correct at least).
While we're at it let's compare these to the 680/700 spread of which we can sell 35!!! ($70k / $2k). The 680 put is at $4.90, so the 680/700 spread has a credit of 8.15 - 4.90 = 3.25. $325 * 35 = $11,375.
Comparing these three positions - I can use a $70k cash reservation for the next week to either earn $815, $4445, or $11,375 (assuming these go to expiration OTM). That's the reward / benefit side.
The risk side is that a full loss needs a share price move to $0, $600, or $680.
This is why I say this is leverage. And at least when I say leverage that's shorthand for "earning or losing money faster". Earning faster is good - losing faster is bad. So like fire - treat it with respect; make sure you have backup plans to handle losing situations, and ideally backups to the backups.
Personally - that insurance put at $680 plus using all of the leverage looks way too dangerous for my taste. And more important - if we're doing these every week for a year, max losses will occur. With the reduced management window, max losses will be more common. The higher credits make the max losses less painful ($59k vs $66k) though...
I've also evolved and decided that a 650/700 ($50 spread size) is also too risky for me for the same reasons. There is more time from max gain to loss, and more effective management window, but we've also been seeing $50 daily share price moves within this year. They're becoming increasingly rare but still, I'm looking for income, and an important characteristic of income is low volatility and regular results.
Actually - that $4.4k outcome using $70k backing is over 5% on the position; 2 of these ($140k backing) per week nearly accomplishes my own income objective ($10k/week) - and without selling any calls which I also do. So I don't -need- to be any more aggressive (use more leverage). This is where having your own target comes in - somewhere soon after objective is accomplished, all of my incremental thinking and resources goes into lowering risk.
With extra resources one of the first things I do is put on more contracts and go further OTM at the start of the position.
