- Dynamic position sizing of 3xIM when market is in an uptrend and VIX at calm levels (< 20). Reducing position by half when in sideways mode. Staying out when in downtrend (like right now). This thinking is in the same lines of pressing bets when it's safer and reducing risk when it's not so safe. By uptrend I mean higher highs and higher lows, sideways is where it's not clear, and downtrend lower highs and lower lows.
- During downtrend when I am not selling ES options, I am going to look at other markets that are calmer to conduct business until things settle down.

I have been playing around with different combinations of spreads for ES to try and minimize black swan damage and came to the following conclusion:

- Less risk means less profits.

Not always. Preventing one 30+% loss can make up for slightly less profit per trade.

- Minimizing risk can be achieved through proper position sizing.

Position sizing would not have prevented major losses last week.

- Complex multi-leg spreads are difficult to execute (and unwind) and I will inevitably screw something up royally (which is another type of risk).

Buying 2 longs for each short isn't that complex.

- Can't eliminate risk completely. We are getting paid to take risk.

Agree

So with these new revelations, I have decided that I am going to use the following:

- Dynamic position sizing of 3xIM when market is in an uptrend and VIX at calm levels (< 20). Reducing position by half when in sideways mode. Staying out when in downtrend (like right now). This thinking is in the same lines of pressing bets when it's safer and reducing risk when it's not so safe. By uptrend I mean higher highs and higher lows, sideways is where it's not clear, and downtrend lower highs and lower lows.

Do a backtest and pretend it was 8/17/15. What positions would you have had on based on the information of that day? Where would those positions be 7 days later.

- During downtrend when I am not selling ES options, I am going to look at other markets that are calmer to conduct business until things settle down.

I understand that this adds a different kind of complexity to the strategy, such as being able to interpret the daily chart correctly. But my thoughts are that using a mutually exclusive element to this strategy could provide a dimension and insight that can't be gained from just looking at the numbers. Of course this is all easier said than done and I have no way to backtest or provide quantitative numbers to back this kind of thinking, but @ron99 has contributed so much by sharing his strategy and want to contribute as much as I can... even if this strategy has glaring holes.

Please feel free to provide feedback as brutal as it may be.

Question is how accurately can you guage when you are in an up-trending/sideways and down-trending market? On Aug 19th I think you would have considered us to be in a sideways market but just 3 days later we traded almost 11% lower and the VIX went from 15 to over 40. Also while I understand that there is a reason the VIX is called the 'Fear Index' and that high levels often indicate uncertainty, but selling options in higher volatility environments means a lot more premium, which means a lot more protection from both a break-even price and 3xIM standpoint. Prices screamed last Monday more due to vol than price.

Do a backtest and pretend it was 8/17/15. What positions would you have had on based on the information of that day? Where would those positions be 7 days later.

I sold some puts with the volatility increase but it was a super small position.

How'd they do?

Term structure and skew can help you make a better R:R trade, or perhaps keep you out of a dangerous market altogether.

They're doing fine so far. I sold the OESZ5 P1100 for 4.00 on 8/25 which was about a .03 delta-ish at the time. It settled at 2.20 today.

@CafeGrande What do you mean by term structure? I've never heard of that term before and I would appreciate it if you could expand on your comment.

Term (time) structure of the market. Because a lot of the big money speculation takes place in the front months you'll often see a steeper IV nearby than in the deferred months (this may change depending on product, seasonality and sentiment). If I remember correctly, the VIX method captures 8 to 30 day options and you're quite likely selling farther out than that. Thus, it pays to look at what you're actually going to be selling, its history, and what the structure might mean. Is it out of whack?

Here are two examples.
1. Gray and white background. Standard and serial ES options (no weekly, no end of month)

2. Tan background. "Constant maturity" options comparing the term structure on Mon 8/24 with Tue 9/1. Note the very steep structure for 1 week and 1 month options on Mon the 24th (and this was at settlement, after most of the craziness) compared to the Tue 9/1 close.

You can actually go a step further and model term structure as a 3D surface -adding in Strike so then you are also modelling volatility skew as well as volatility relative to expiry. A lot of the commercial option software packages will do it for you but it's pretty easy to do yourself in excel.