i am thinking of black swan, especially for 2022... therefore, i am analyzing the benefits of BPS width vs straight puts...
assuming $2M capital for 12/23:
-p950 x50 credit $115k, width 0
-p950/+p900 x500 credit $487k, width 50
-p950/+p850 x250 credit $379k, width 100
-p950/+p800 x175 credit $322k, width 150
-p950/+p750 x150 credit $300k, width 200
-p950/+p700 x100 credit $205k, width 250
-p950/+p650 x75 credit $159k, width 300
if i want less risk and i prefer capital preservation instead of more income, the lowest trading risk is just straight up sell puts instead of BPS (regardless of width), amiright? There is no way stock will drop to zero. Capital is gone on BPS max loss.
thoughts? comments? objections? analysis?
TIA!
I see that some of the numbers are off and that leads to the credits received being off. I suspect its because you figured out how many of each you could open based on margin usage - I agree with the conclusion even though I analyze assuming no margin (and then crank in the margin later - reasonable and successful people will approach this part differently).
For least risk and capital preservation, fully cash secured puts can't be beat. A full loss will require that the shares go to $0. If you're using margin to sell 50 of these instead of 22 (call it 25 for easy math), then a full loss is available at a share price of about $525 (1/2 of current).
The cash secured puts also provide the widest range of management options from the perma-roll to conversion to put spreads later.
But I like the bit more income that can be achieved using put spreads and (apparently) like you if I have $2M cash then I will put it all to work. If I were using $100 wide spreads then I'd open 200 of them. If I were doing $50 wide spreads then I'd open 400 of them. I mitigate this tendency of mine by using wider spreads - I use $300 wide spreads right now, and thus would 'hold back' and only open 65(ish) of them. Fewer contracts = less leverage = losses accumulate more slowly as the share price goes down.
A side benefit - if I were to open $100 wide spreads instead of $300 wide spreads and hold back 2/3rds of my money so I don't have it all at risk then my income is better with the $300 wide spread (insurance put is cheaper; contract counts are the same). Of course I can also lose 3x the money if I sit back and do nothing or the shares crash really badly. Except my losses would actually be the same - I'd have 3x the contracts on the $100 wide spreads (not-advice; that's how I'd do things, not necessarily because its smart).
The max loss is the same in all of these positions ($2M - credit received) - just that max loss arrives much later with the wide spreads over narrower spreads, and it arrives the latest on naked puts. And I can open these positions in a US retirement account that is otherwise restricted to CSP.
I did the analysis below a few days ago so the premiums will be out of date. However the main point of the exercise was to look at the premium received versus max loss risk for various spread sizes. Unlike what I'd previously thought, the wider spreads provide a similar premium for the (portfolio) margin used, sometimes slightly better. However the max loss dramatically increases with wider spread width. So if you are giving a higher priority to mitigating a black swan then a narrower spread width is preferable. The straight Put still results in a large max loss, you are just selling less of them.
Spread Size | Option | Number | Premium | Revenue | MM Used USD | Max Loss USD |
$30 | 1150/1180 BCS | 121 | $1.88 | $22,748 | -$100,000 | $284,500 |
$50 | 1150/1200 BCS | 91 | $2.62 | $23,804 | -$100,000 | $476,700 |
$100 | 1150/1250 BCS | 69 | $3.57 | $24,743 | -$100,000 | $967,199 |
$200 | 1150/1350 BCS | 61 | $4.15 | $25,379 | -$100,000 | $1,961,999 |
$30 | 820/850 BPS | 152 | $1.67 | $25,348 | $100,000 | $288,000 |
$50 | 800/850 BPS | 105 | $2.40 | $25,166 | $100,000 | $479,000 |
$100 | 750/850 BPS | 75 | $3.56 | $26,539 | $100,000 | $967,199 |
$200 | 650/850 BPS | 63 | $4.78 | $29,996 | $100,000 | $1,954,499 |
This is great info and important to understand about how portfolio margin differs from regular margin, at least on US brokerage accounts (is that Reg-T margin?), and further differs from US retirement account trading restrictions. Considering just how very different these numbers are from what I have available and am doing, these effectively make for an entirely different analysis.
With roughly immediate study on my part, I'm pretty sure that my whole analysis and thinking about what and how to do would get to be redone from scratch
Can you guys please chime in with your opinion regarding wide or narrow width spreads?
It appears that narrow spreads are less risky than wide ones. For example, for 12/17:
-900p/+700p - max return $587, max risk: $19,414, prob of profit: 93.9%
but
-920p/+900p - max return $210, max risk $1,790, prob of profit: 90.1%
So, with the wide spread I'm risking $19k for $587 but with the narrow one I'm only risking $1.7k for $210 with pretty much same probability of success. If I sell 2 narrow spreads, I'll be getting about the same premium but risking less than 5 times the loss. Even if price is below $900, causing max loss on the narrow spread, the wide spread will produce $586, so, outperforming the narrow by $586+$1,790=2,376 and by about $876 stock price, only $24 less, wide spread will lose more money.
Wouldn't it be more statistically profitable to sell narrow spreads at about the same distance as wide and risk much less capital for similar premiums? Even if manageability of narrow spreads precludes them from rolling, the potential for large losses is much less than wide spreads given the far distance from current price
Spreads are a form of leverage, enabling us to take on larger positions than we can achieve using straight up cash (cash secured puts). They're also leverage that is available in retirement accounts, where leverage in the form of margin is not available in retirement accounts (US style anyway).
What makes the $20 wide spreads particularly risky is the tendency for some of us (me!) to use all of the available money. So if I were choosing between 1 of the $200 wide spreads and comparing its results to 1 of the $20 wide spreads, then I'd earn more premium in exchange for 10x max loss. Definitely not 10x on the premium side.
But that isn't what I'd be doing - I'd sell 1 of the $200 wide spreads or 10 of the $20 wide spreads. Now my capital at risk is identical - the difference is that max loss happens $20 ITM instead of $200 ITM.
Your analysis is entirely spot on and if you have the $20k to back one $200 wide spread or would choose to sell 2x $20 wide spreads, then yeah - same premium and a lot lower max loss. My personal problem in that situation is I'd be like "yeah - I can sell another 8 of these small spreads and earn 5x the premium!@%(&!". And then I'd do it, and then the insurance put would go ITM and I'd lose the $20k minus the collected premium (which will be larger on the 10x$20 wide spreads, but still).
So I use the $300 wide spreads for a few related reasons. First I do want some leverage in my positions - there's a big increase in income from using that leverage and I like that. However that also means I'm pulling my max loss strike up from $0 to my insurance put strike (assuming no management, the insurance put goes ITM, and I hold to expiration). In that 950/650 example of
@Yoona shares need to reach 650 for a full loss, so that's part of my risk management - I'm a long ways away from a max loss.
From what I've seen these wide spreads behave almost identically to naked puts for some distance ITM on the short put. I haven't studied it intensely but I know that at 1/2 of the spread the available rolls are straight out in time for ~$0 debit/credit. In practice the rolls will be getting to look like that in the vicinity of the 1/2 way point. Below the mid point you'll be paying a debit for a straight roll, or rolling further ITM to get a credit. (Ugh). They get worse - it wouldn't matter if you were rolling 1 week or 3 months when you're that far ITM - it'll be the same dynamic for 1 or 10 week rolls as the midpoint in the spread is where time value on the two options is equal.
As best I can tell, up to about 1/4th of the way ITM the position is still behaving more like a naked put than not. Reasonable rolls are available that will improve the strike (maintaining the spread width) for a credit. I'm trying that out right now - I rolled 750/1050s for this week to 735/1035s for next week while the shares were at $970-975 (the 1/4th point ITM). These numbers are VERY dependent on the IV - over the summer IV was low enough that $40ish ITM was enough for a naked put to be into the rolling straight out for minimal credit land.
Anyway - with the wide spreads, the 1/4th point is big enough to provide some room to maneuver. On a $50 wide spread, those $1050 strike puts would have seen a max loss show up, at least for a day or two. And the good management rolls would be available down to ~$1038 with minimally effective rolls down to $1025; almost nothing considering how fast TSLA moves around.
Not directly in the discussion but applicable to me at least and controlling my tendency to use everything available, my overall account management target is to use at most 1/3rd of the account value as cash to back put spreads. The other 2/3rds are used to purchase long calls (usually 6 month DTE in retirement accounts and 24 months DTE in brokerage - a topic for another time) and use those calls to sell covered calls.
I don't do call spreads (personal choice) and I treat the 2/3rds in purchased calls as cash I would use in the event of a significant loss of cash. That'll keep my cash devoted to put spreads reasonably constant, and allows me to experience 2 max losses close together and still maintain the same level of put spread sales.
If I tried to hold that 2/3rds as cash, or 1/3rd in use, 1/3rd as cash not used, and 1/3rd in purchased calls, I would be looking at that 1/3rd unused cash and figuring out what I should do with it