buy or sell question - 28 months analysis

so here would be my question on this:

lets say to minimize left tail risk you buy deep out of the money put only. you are only protecting against a meltdown or 9/11 type event. but instead of monthly you several months out so the premium that you pay gets you three or four months of coverage. how does this hurt profitablility?

seems like there would be a little more art than science to this but then again there is always someone that can optimize pretty much everything
 
Quote from h hubbins:

so here would be my question on this:

lets say to minimize left tail risk you buy deep out of the money put only. you are only protecting against a meltdown or 9/11 type event. but instead of monthly you several months out so the premium that you pay gets you three or four months of coverage. how does this hurt profitablility?

seems like there would be a little more art than science to this but then again there is always someone that can optimize pretty much everything
More...

If you use the thinkorswim "thinkback" function you can backtest to see how buying a DITM put as insurance works in various scenarios. For example, you could have purchased a DIA June08 100 put on 1 Oct 07 for $61 (ask). With the drop in DIA stock price, your "insurance" would have gone up in value $95 as of 10 Mar. On 11 Mar the put contract profit would have slid to $37.50.

Bottomline is the put can either augment your profits, offset your straddle/strangle gains, or add to your straddle/strangle losses.
 
thanks for the tip on the think or swim software. i've been selling some strangles and just winging it with the out of the money puts since there is some major systematic risk to the market now. so far so good but that software might help.
 
I should have said DOTM puts rather than DITM puts in previous posting.

Something else to do in addition to looking at the historical price changes is to analyze how increased volatility will effect both the strangle/straddle and your protective put. Crunch the numbers and see how puts at various strike prices react as the volatility jumps up to various levels. It's not as simple as just buying a DOTM that is 20 or 30% below the current strike price.

Perhaps you could build on the existing spreadsheet on this thread and share the results with all.

Best of luck.
 
Quick update and new test results.
1. My spy selling call and put for March 2008 for $8.2 strike 135 is still $1.7 ahead (even with all the down movement).
2. I repeated the analysis for last 28 months for qqqq, djx, and spy and to simplifiy the strategy I sell put/call for the closest strike to price (and I make all the fields in the excel calculated fields – to eliminate my error) so excel should be easy to read by those who follow.
3.
Observations:
Looks like the more expensive index the better results (percentage wise, as percentage is: what percent of initial credit I am keeping after expiration. Maximum would be %100, and loss is unlimited.)

While buying deep OTM puts provides protection but it erodes the profit.
After some analysis the only good protection is (obvious money management): close the position if the loss is x% (I used 0.75 as x). Namely, if for march 2008 I collected 8.2, so I would close the position if put/call cost: 8.2 + 0.75*8.2 = 14.35.

Lehman Brothers (are they authority anymore) person (from: http://www.bfinance.co.uk/inst/article.do?serieId=1&docid=N12504) mentioned:
“, if done on a systematic basis, this is profitable”.

I could just speculate that a lot of funds, firms, etc., are using similar strategies to collect: “Volatility Risk Premium“
Attached excel had data for djx, qqqq, and spy.

I will attached csv (zipped all strikes, months) quotes for qqqq, spy, and djx for 28 days (expiration days) if anyone wants to play around.
 

Attachments

Mark -

do you have the data to determine how often you would have touched your stop-loss rather than whether the expiration date price would have hit the stop-loss? I imagine quite a few of the profitable trades would have been closed out with your 75% rule.

Also, if you've already crunched the number for the OTM puts, would you mind posting that data?
Thanks
 
1. Protection to the downside by buying puts so that there is almost no downside risk (strike of puts = strike for selling puts/calls - initial credit for puts and calls, eg. for March 2008 we sold 135 strike for 8.2 so the protection puts would be at 127). The average gain is reduced to 10% and the puts were useful 3 times in 28 months.
Attached is a zip with djx, qqqq, and spy options quotes for 28 recent expiration Fridays so you could play around
 

Attachments

I found your short straddle data intriguing, so I took a closer look at some things. First off, the losing months typically occurred during transition periods for the VIX. As the VIX moved from low levels to high levels, the straddle lost. As VIX moved from high levels to low levels, the straddle lost. But, using a stop loss prevented serious losses (whatever you view as serious). Now, the gains are typically 50% of the initial premium received. Therefore, the risk needs to be closer to 50% than to 75%. If we compromised and used 60%, then your big losses (over 71%) would be avoided. Now, someone mentioned touching the stop and bouncing back. I don't think that happens often. After some research, it seems that even if the market moves 100 points in either direction (down is worst), the loss is only around 50% anyway. The Mar08 straddle is a great example. The market (SPX) did move 100 points down from the entry point and is now within 19 points of the entry point. Also, one should adjust the number of contracts, to reduce risk, when the total premium increases and decreases. If I were trading for a living, I would have used the ES instead of the SPY options (the numbers are the same, just get more bang for the buck with ES options). Second, during 2006, I would have needed to use 10-12 contracts per straddle. In 2007, I would have reduced that number to 8-10. And looking at this year, I would use about 3.

Additional Info: When the VIX was low (between 10 and 15), I simply sold puts about 3% OTM and sat on them until they expired. I had about two losses a year in 2005 and 2006. In 2007, I began to hedge the short puts with short futures. Losses would have been managable if I followed my rules during the transition period from low to high vol. The problem with that technique in a high volatiltiy situation is that the futures bounce around too much and I have to take losses in them as the hedge needs to come off (delta becomes <.075). Your short straddle strategy seems to do well in high volatility situations and in low volatility situations. It doesn't do well in the transition periods from low to high or high to low, so, maybe a change is strategy is required during those periods.
 
jwcapital,
I agree with you in most of the points and I am glad someone took time to confirm my ‘findings’ and apparently used it in practice.
Points I want to make:
1. I realized quickly that we need to adjust the number of contracts and this was the reason I used percentage (or fraction) of the initial credit we keep (or loose) to be more objective (this month it was $8.2 with high vol. but other times it was around 4 with low vol.)
2. Transition from low to high vol. is hard to predict, after doing some more testing I discovered that by automatically closing the position (without any adjustment) with 10-30% loss still gives nice returns over 28 months.
SPY in March just collapsed but it looks we still are going to have better than average profit (break even with SPY around 127).
Will share more of my findings.
 
Quote from markg_ny:

jwcapital,
I agree with you in most of the points and I am glad someone took time to confirm my ‘findings’ and apparently used it in practice.
Points I want to make:
1. I realized quickly that we need to adjust the number of contracts and this was the reason I used percentage (or fraction) of the initial credit we keep (or loose) to be more objective (this month it was $8.2 with high vol. but other times it was around 4 with low vol.)
2. Transition from low to high vol. is hard to predict, after doing some more testing I discovered that by automatically closing the position (without any adjustment) with 10-30% loss still gives nice returns over 28 months.

This is interesting. It was funny you initially picked 75% loss. Your best loss, according to my calculations was 41%; your worst loss was 153%. If the market has 100 point swings, you will stop yourself out a number of times. It is not unreasonable to have a risk:reward ratio of 1:1 (50% gain vs 50% loss; as I mentioned, your typical gain is 50% of the total, initial premium. Are you saying that by getting out with a 30% loss that I won't miss gains even when the market swings 100 points in one or both directions? I ask, because with a range of 10-30%, you would really have an incredible risk:reward ratio. How many of the listed gains would be missed if stopped out because of a narrower %loss?

SPY in March just collapsed but it looks we still are going to have better than average profit (break even with SPY around 127).
Will share more of my findings.
More...

As of 11:40AM ET, the total premium of the Mar 1350 Call and Put is 77.50. So, since your initial total premium was 82, your profit is 4.50 per straddle of one contract. In ES terms, this is a $225.00 profit. While it isn't enough to live on, it is still better than a loss in this market. When the market dropped to 1255, your loss would have been about 50%, and you would have been stopped out based on your 10-30% stop loss limit. I think in this environment, you need to give the short straddle some room to breath--that is why I suggested 60%. If I traded the Mar08 1350 put and call short straddle, the total premium would be 82.00 per single-contract straddle. Based on your data, my profit expectation would be 41.00 per single-contract straddle ($2,050.00). My loss would be 60% (based on my stop-loss suggestion) or about a 48 point loss ($2,400.00). That translates to a risk:reward expectation of about 1:0.85. A number of traders would not approve of this ratio. You gotta believe your expectation for success is higher for this to be a good trade. A 60% loss eliminated your three biggest losses over the 28-month period (losses were 71%, 107%, and 153%). Your next highest low was 54%; that is why I choose 60%; to give the trade room to breath. Please continue expressing your thoughts in this thread.
 
You must log in or register to reply here.
Back
Top