- collect more premium per unit of trade or per commission dollar
- collect more premium before decay starts to accelerate
- one can ride out short term technical moves against a longer term functional view
- one can sell much farther out of the money options to lower risk of going in the money
- one will enter the last 90 days at price much farther out than someone who waited to sell
- less stress for the above reasons
- lower probability of going in the money at most points during trade
In short, the pitch is that it is a type of free lunch (a phrase that is never used by anyone; this is just me juicing the discussion): more money for less risk. Sounds magical but I'm not totally convinced after doing this for a few years. The market only gives what it will give and you can only sell what someone will buy. Sometimes that does not translate into less risk, just different risk and the risk you really want is not available in the market at all. Again, I think they are right and for the right reasons. I just have a hard time pulling it off in practice.
Here is an example using soybean futures. Shorter term options (as of 10/13) weren't offering me much so I nudged myself out to January (Dec 23 options expiry) or close to 2 and a half months of tenure. I should have gone further but didn't. When I look at what I can sell and the approximate probabilities, what I see is that yes I am further out of the money but I just don't feel like my risk for what I'm getting is all that much less since with more time there is more room to run. It's simply a function of volatility and time (ignoring functional trends and seasonal tendencies). Also ignored is the possibility of major shocks; status quo in vol is assumed. Also the chart itself tells a story that is not all that reassuring. This is how I look at it.
1. The approximate probabilities for price movement tell me that I don't get very far out in standard deviations of possible price movement before I lose a ton of premium (but I already knew that):
- The red bars are an amateur hack. It is a premium intensity function of current option prices. It approximates, but is not, a probability distribution of price but it does show the skew to the right based on what the option market is doing/expecting. This compares favorably to the Interactive Brokers Probability Lab which I cannot yet replicate or replace on my own and which is not displayed here.
- The red dotted line is the price range where option premium per contract is > $100.
- The green dotted line is the price range where option deltas are < 20 (for the sake of argument).
- The little red and yellow circles are the one and two standard deviation points for the two distributions.
So, if I want to be at least one stdv out (all else equal) and also get more than $100 and also have a delta somewhere around 20 or less, then the best I can do is something like go short a put at 890 and/or a call at 1040. That's if I want to manage my risk carefully and still make some money. But it's not that much. If I got it right that's about $275 each or $550 together. I mean, that's ok but then I look at the chart and I'm a little dismayed.
2. The chart tells me that for the 2.4 months I am considering, I can get totally smoked if I do not have a risk management plan in place, which I would have if I put on a trade since the risk return is seriously upside down.
In other words, going to "their" 2-to-6 month sweet spot (I should have gone further out than 2.4 but I expect it to look similar at 3-4 months), it does not look all that sweet. It does not look less risky unless maybe one assumes that there are no dramatic moves in the first x weeks of the trade (after which one might be able to relax a bit). This is a sotto voce but critical assumption mentioned by the pros. It's one that needs to be turned up quite a bit in volume, though. Here, this is a little unscientific, but this is what I see when I look at the daily chart for January delivery.
- Yellow on the right is the expected tenure range. The other yellows are just that range projected onto the past price data.
- White line on the left is the price move over the expected tenure that was projected onto the past data. White on the right, then, are just a copy/paste of the white lines on the left but now projected into the future over the time frame of the expected option tenure.
- The blue lines, if you can see them, are the option strikes of 890 and 1040, or close enough.
Concluding thoughts?
No, not really. This post is more of a question than an answer; it's more me working this out in my own head as a process than anything really worthy to dwell on. Keep in mind I am still a journeyman on all this and I'm sure I could be easily beaten down by someone more expert than I am. It's just that I find myself not all that dazzled by the premium/risk enhancements gained by going out a bit and the pro's stuff that I was reading, while convincing on the page, is not as convincing when I look at the market. I'm sure it's there and I will continue to look but it is definitely not a free lunch.
Postscript
I was thinking about it some more and I guess the only point I was trying to make was that these guys (and we too) can't predict the future so if one has only price, volatility and time go on [hold that thought for a second] then necessarily the probability distribution expands the further out you go in time. That means you can go farther out of the money going further out in time but the core probability risk will go with you, all else but time being equal. This is not genius, it's more like high school math and looks like this:
And, to give the guys I'm talking about some credit I don't think they really are missing that point or at least I hope not. If you look closely at the book and site that I'm thinking of, they will make an assumption a priori [back to that thought I had you hold a moment ago] that one has a functional point of view on price BEFORE the analysis is begun. So where they are right is that the further out of the money one goes, the farther one can go beyond (or towards) one's point of view on some "threshold point." The example for this might be better in crude oil. With supplies high and a ton of supply poised to come online with any uptick in price and with OPEC and Russia and Iran not playing well despite rumblings about agreements and with demand having been in the doldrums for a bit, it wouldn't be unreasonable for any given trader to have a POV that price won't go above 55 or 60 for a while. That means that if one were to sell an option above 55 or 60, which I did btw, the probability of the trade expiring out of the money is goosed a little, in my opinion. Not certain, just goosed. That argument I can understand and support. I'd withdraw the entire post since I more or less just mooted myself, but why bother? (So far only my sister and one other person reads this stuff as far as I can tell,)
No comments:
Post a Comment