Weekly RUT IC

[cwelsh 1,497]

Please see my earlier post for the risks of these trades (its high):

Just sold the following on the RUT Jun 29 weekly:

720/730/795/805 credit of .55, margin $5.00 -- after commision return of 10.5% if it expires worthless (which it should).

RUT is at 762, so it would take a 32 point drop or 33 point rise to actually lose. Though do note that if you get a 15-20 point rise (or drop) in the next two days the trade will reflect a book loss.

The idea is to let this one expire worthless so you don't have commisions closing it out. If it moves adversely against you, it likely will do so quickly. If you're ok with a 5% return, then you can probably get out in two days.

Will probably enter the Aug DOTM IC next week when the spreads close some.

[Kim 7,423]

What do you mean by DITM IC?

[cwelsh 1,497]

What do you mean by DITM IC?

Typo should have been DOTM -- (deep out of the money) -- already fixed, thanks.

[Marco 221]

I was back testing the idea of selling weekly RUT IC's on a weekly basis and was looking at what distance the strikes to ATM should have (close to ATM = high premium but also high(er) probability of loss, wise strikes = low premium and therefore high loss compared to initial credit but much lower probability of it happening) I actually found ~4.3% (the distance you choose for the short strikes) to be a good compromise between the two, however I was looking to open the trade on Thursdays and receiving ~1$ credit for a 5$ spread! Even with that credit the trade wasn't immensely profitable on the long run as in volatile times RUT regularly moved more than 4.3% in a week so I was looking for a 'filter' that would stop me doing the trade in volatile times but get me back in quick enough to make up for some losses (you only realise its a volatile market once you are in it so you will have lost on the trade at least once) while Implied vol is still high. I think (need to double check when Im back from holiday and have access to my spreadsheet) I ended up with the following: If the 6 week rolling STDEV of WEEKLY returns on the RUT is over 3 I would stop the trade and get back in once it drops below that.

Back testing that, assuming 1$ credit (or the respective %equivalent premium going back, which is a big question if that's reasonable to assume) didn't look that bad however you still had quite bad draw downs at some periods which stopped me from implementing the strategy as I couldn't allocate the size I would have liked to and I decided to stick with the metholgy that Kim uses (enter about 6w ahead of expiry, be out at about 3w to maturity and adjust strikes if necessary) that way I think you can manage loses much better and can throw more money at the trade.

I'd be interested if you found a 'solution to the problem' but I don't like the risk/reward * probability of this trade - the potential max. loss of 10$ would wipe out the profits of 17 max profits trades. Do these move occur much less that 5.5% of the time? I can't test it where I am now but from what I learned over my back testing I doubt that this would be a good trade to do every week and even stopping it at volatile times....

[Mark Wolfinger 14]

720/730/795/805 credit of .55, margin $5.00 -- after commision return of 10.5% if it expires worthless (which it should).

Margin is $10.00 for the 10-point spread

RUT is at 762, so it would take a 32 point drop or 33 point rise to actually lose. Though do note that if you get a 15-20 point rise (or drop) in the next two days the trade will reflect a book loss. Wow. A 'book loss' is a loss. It may not be a realized loss (yet), and that loss may disappear. However, traders who ignore future risk and believe that 'book losses' are not real will suffer the consequences later.

Please be aware of risk at all times. Please be aware that iron condors do not win every time. The path to long-term profitability is knowing when to accept that 'book loss' as real.

The idea is to let this one expire worthless so you don't have commisions closing it out. If it moves adversely against you, it likely will do so quickly. If you're ok with a 5% return, then you can probably get out in two days. Accepting the additional risk of holding through expiration just to save commissions is another mindset that you will have to change to find success as a trader.

Are you saying that you are not ok with a 5% return in two days? When Gordon Gekko said that 'greed is good' he did was not refering to this type of wager.

Good trading to you

Mark

http://www.mdwoptions.com/Premium/home-page/

[Mark Wolfinger 14]

I was back testing the idea of selling weekly RUT IC's on a weekly basis and was looking at what distance the strikes to ATM should have (close to ATM = high premium but also high(er) probability of loss, wise strikes = low premium and therefore high loss compared to initial credit but much lower probability of it happening)

ICs with options that are closer to the money do not have a higher probability of loss. They have a higher probability of moving into the money. These are not the same. In fact, with proper risk managment these CTM condors provide better long-term returns. But - that depends 100% on the way the trader handles risk.

If the 6 week rolling STDEV of WEEKLY returns on the RUT is over 3 I would stop the trade and get back in once it drops below that.

Whatever method you use to determine 'too volatile,' it is a very good practice. There is no necessity to trade these positions every week.

Back testing...you still had quite bad draw downs...

Test again by opening positons on Mondy and avoiding those Monday gap openings.

I'd be interested if you found a 'solution to the problem' but I don't like the risk/reward * probability of this trade - the potential max. loss of 10$ would wipe out the profits of 17 max profits trades...

The largest problem is that you do not seem to allow for the possibility of rolling dowm, reducing size and holding, or exiting early. Condor trading should not be all or nothing. Risk management is th key to sucess.

Regards,

Mark

http://www.mdwoptions.com/Premium/home-page/

[Kim 7,423]

Thank you Mark,

I think that taking a settlement risk is a bad practice. Trying to save on commissions on those trades (which are not commissions consuming to begin with) is not a good idea. I'm not sure if I would roll the weeklies, but exiting early is definitely a better idea, once you have 5-7% gain. I also agree that entering on Monday might be a better idea since weekends tend to produce big news many times and the risk of big move on Monday is fairly high.

The key here is to find a balance between exiting too early and letting a small loss to turn to a large loss. I think with weeklies, it might be much more difficult than with "traditional" ICs.

[Marco 221]

That was one reason for me to drop the idea of weekly IC's - it's going to be much harder to adjust them if things go wrong as due to the short maturity things will go wrong fast and you are very quickly from 'everything still according to plan' (a say 15-20 point move) to your max loss.

Also with only 55c credit to start with its going to be hard to roll strikes after a 15-20 point move and maintain a credit. And if you already roll after RUT moves half the distance to your short strikes you might as well pick a closer distance and get more credit upfront.

While I agree on don't be shy to close out the trade early and 'only' take say 70% of your profits, I think taking a 25c-30c profit on a trade where you risk a 10$ loss (even though if it's only ~5% probability) makes the risk/reward ratio even worse for me.

That's still the major draw down for me with this low credit trades - that you can't allocate serious money (as % of your equity) to this trade as a 2 STDEV event will wipe out too much of your portfolio. With the 6w version, a 4.5-5$ credit to start (for the 10$ spread), possibility of a strike adjustment and closing it way ahead of maturity you have a good chance to make 15-30% profit depending on how IV and spot behave and more importantly you have an excellent chance to limit losses to ~30% in a bad case scenario. This way you can take much bigger positions and end up with higher profits (in $ and % of your portfolio) than with the weekly version.

[Kim 7,423]

I completely agree Marco. And if you combine those 6-7week ICs with the earnings plays, you have a perfect balance.

[Marco 221]

I completely agree Marco. And if you combine those 6-7week ICs with the earnings plays, you have a perfect balance.

Yep exactly, and even there the '6w version' works better as it will require a much larger move (10-12%+ in a short time) to seriously derail that trade whereas a say 6% in a week would get you in much more trouble on the weekly trade. In the first case you'll have a lot more fun with your earnings trades.

[Rogers 263]

IMHO Chris has his ducks in a row and understands the high risks involved.

As many have mentioned before it's about knowing your risk tolerance and managing those risks appropriately. I enjoy reading about his various trades and the thoughts behind them.. Most of them are out of my comfort zone but I have tried a few and been happy with the results.

[cwelsh 1,497]

720/730/795/805 credit of .55, margin $5.00 -- after commision return of 10.5% if it expires worthless (which it should).

Margin is $10.00 for the 10-point spread -- you are correct, was looking at 5 point spreads, but couldn't do a five point spread on the the put side.

RUT is at 762, so it would take a 32 point drop or 33 point rise to actually lose. Though do note that if you get a 15-20 point rise (or drop) in the next two days the trade will reflect a book loss. Wow. A 'book loss' is a loss. It may not be a realized loss (yet), and that loss may disappear. However, traders who ignore future risk and believe that 'book losses' are not real will suffer the consequences later. -- I believe our symantics are different here. If RUT drops 10 points tomorrow, this will reflect an unrealized loss -- however it could very well be within the expected deviations and even possibly be a better position for the trade overall. The trade could reflect a 5% unrealized loss but still have an expected positive return. At no point am I ever ignoring future risks -- please go see my AMZN/GS posts for a much larger warning on the risks of this type of trade -- they are carefully positionally sized, don't take up a large portion of the portfolio, and have a fairly large probability of being succesful over the longer term if I stay within my trading guidelines (they have been for three years anyways).

Please be aware of risk at all times. Please be aware that iron condors do not win every time. The path to long-term profitability is knowing when to accept that 'book loss' as real. No kidding -- you can easily suffer a 50-90% loss to margin on one trade like this. I have very set exit points (well set is again poor word usuage as the exit is dynamic depending on time left to expiration, volatility, and recent SD moves). That said, if over twenty trades you suffer a 50% loss to margin on one trade (-$5.00) and a 25% loss to margin on another (-$2.50), and profit fully on the other 18, that's a very nice return.

The idea is to let this one expire worthless so you don't have commisions closing it out. If it moves adversely against you, it likely will do so quickly. If you're ok with a 5% return, then you can probably get out in two days. Accepting the additional risk of holding through expiration just to save commissions is another mindset that you will have to change to find success as a trader. I 100% completely disagree -- I think this depends entirely on the trade. If RUT is within a few points of the strike, of course you close it out. In reality you probably should have closed before it ever gets close to that point. But if RUT is right at 762 thirty minutes before close, I would almost never close it out (might if there were major fed annoucements or something along those lines). Yes you should always close out trades that have a large degree of risk, but when the risk is minimal, on transaction cost heavy trades, I don't think you should (and I've tracked this as well).

Are you saying that you are not ok with a 5% return in two days? When Gordon Gekko said that 'greed is good' he did was not refering to this type of wager. Of course I'm happy with that return, though over a year my expected return on these trades is just north of 1% per five day holding period, not 5%. That's due to taking losses when it gets above or below my risk points, or having sudden adverse moves that I have to dive out of and end up with a large loss -- but again that large loss is expected and accounted for.

Good trading to you

I appreciate your comments, and maybe we should chat about overall portfolio allocation sometime. Just for your piece of mind on my trading, I run a full monte carlo on my portfolio up to 5 standard deviation moves out each weekend, so I keep a pretty good idea on overall portfolio risk/reward.

Specifically, on the weekly RUT ICs, I only trade them about 15-20% of the time -- they have to meet certain perimeters. The two larges are that a trade must be both (1) outside of two standard deviations over the last 14 days and (2) to be at least as far out as the largest move RUT has had in the last 30 days over the same time periods (so in a five day, what is the largest five day move in the RUT over the last thirty days?). I cacluate the SD by taking the last fourteen days of data, tracking the 1, 5, and 7 day movements, and calculating the SD moves there. In this case the trade was even better, because it was outside of two SD's using the last 14 days of data and last 30. I use a similar method in calculating exits -- so that's more of a dynamic calculation.

In backtesting, if both those conditions are met, on the RUT (as I stated) I average just north of a 1% return (1.15%).

[cwelsh 1,497]

Yep exactly, and even there the '6w version' works better as it will require a much larger move (10-12%+ in a short time) to seriously derail that trade whereas a say 6% in a week would get you in much more trouble on the weekly trade. In the first case you'll have a lot more fun with your earnings trades.

As to Marco's and Kim's post -- I do prefer the longer dated IC's -- less risk and more room to maneuver. However, i beleive there is a place for the weeklies as well given proper position sizing.

[Marco 221]

Hi Chris a few questions to your methodology please (please understand them as questions rather than criticism, you seem to approach trades in a similar manner that I do but seem to come to different results so I'm curious as to why that is. I do appreciate your contributions here as they often make me think and even if I may disagree with your thesis in the end they have kicked off a thinking process that lead to a trade more than once in the past)

A) when you talk about standard deviation (STDEV) do you talk about STDEV of returns over the maturity of the trade? So the STEDV of weekly returns on a weekly trade STDEV of monthly returns on a monthly trade and so on? You say one setup criteria is that the short strikes must be out of 2 STDEV of the last 14 days. For a 4 day trade (tue to fri (or don't you count Friday?) you only have 3 four day returns in 14 days - I'm not sure if STDEV of 3 data points is a meaningful figure.

B.) I'm not sure if I understand you correctly but are you saying you take a (IC) trade if you can place the short strikes outside a 2 STDEV move and make an average return (incl. losers etc.) of 1% per 5 day period?

A 2 STDEV move happens 5% of the time (per definition) so shouldn't you target 5% plus returns then? (much more actually accounting for the fact that financial markets aren't normal distributed (fat tails, black swans etc.) and for commission, slippage etc) and shouldn't it be independent from holding period as you would adjust your STDEV calculation to your holding period (see question a)

C) I use STDEV a lot looking at the risk and the probability of a return of a trade however not to the extend that you seem to, but I have to say I don't come across too many trades where the market pays you a 10% yield for a trade that needs a 2 STDEV move to make a loss. Even though I have to say I usually look at 1 to 1.5 STDEV especially for shorter dated trades as I don't like selling the small premium of 2 STDEV moves. So maybe the market DOES pay a premium for these however I don't think it's enough over 5% to account for the factors mentioned above. So I don't think you'll beat the market on probability (maybe as a MM receiving the spread rather than paying it) which leads me to question

D) how many of these 2 STDEV trades have you done over the last 3 years? You say you have been profitable doing them but even if you did 1 of these trades a month (so 36 trades) the chances are pretty good (~16%) that you have been 'just lucky' to avoid a 2 STDEV move. You can be 'profitable' quite a long time selling the 5% chance of a big loss for a 2% premium before the odds catch up with you.

Position sizing doesn't help you with that issue by the way as your expected return is still negative over a large number of trades, so while if will prevent you from blowing up your portfolio it doesn't help you to make money if the expected statistical outcome for the trade is negative.

Marco.

[cwelsh 1,497]

Hi Chris a few questions to your methodology please (please understand them as questions rather than criticism, you seem to approach trades in a similar manner that I do but seem to come to different results so I'm curious as to why that is. I do appreciate your contributions here as they often make me think and even if I may disagree with your thesis in the end they have kicked off a thinking process that lead to a trade more than once in the past)

I always welcome thoughts and even justified critiques (heck that's how half my models developed is by other people poking holes in them)

A) when you talk about standard deviation (STDEV) do you talk about STDEV of returns over the maturity of the trade? So the STEDV of weekly returns on a weekly trade STDEV of monthly returns on a monthly trade and so on? You say one setup criteria is that the short strikes must be out of 2 STDEV of the last 14 days. For a 4 day trade (tue to fri (or don't you count Friday?) you only have 3 four day returns in 14 days - I'm not sure if STDEV of 3 data points is a meaningful figure.

Here's the basics -- I import the daily close prices into a spreadsheet in one column. The next column calculates the price changes over a 1 day period (so A2-A1); the next column over a 5 day period (A6-A1); the next over a 7 (A8-A1); the next over a 14 (A15-A1), and continuing for periods over 21, 30, 45, and 60 days. So I now have a list of data points for how much the underlying moves over a certain period.

On page two of the sheet, I have the following:

7 Day

1 day 5 day 7 day 14 day............60 days

Largest move:

Largest absolute value move:

Average move:

Average absolute value move:

Standard deviation: =STDEV(B1:B8) =STDEV(C1:C8)

14 Day

1 day 5 day 7 day 14 day............60 days

Largest move:

Largest absolute value move:

Average move:

Average absolute value move:

Standard deviation: =STDEV(B1:B15) =STDEV(C1:C15)

21, 30....etc,

And then I copy that block 9 times (7 days, 14 days, 30 days, 45 days, 60 days, 90 days, 100 days, 180 days, 360 days)

So the fewest "data points" I have in any calculation is seven. So in my 7 day block above, I know, over the last seven days, what the average one day move is, what the average 7 day move is, and so forth.

Then for trading, I go to the most relevant "box." So on the RUT trade, that I am only going to hold, at max, five days, I would want to know what the average and standard deviation moves are over a 7 day period. But over how big of a time frame -- well for RUT I use the 14 day time frame -- so I have 14 data points. I frequently expand it up to 30 as a way of extra caution, but moving outside of 30, on RUT, has been largely meaningless (via backtesting results).

B.) I'm not sure if I understand you correctly but are you saying you take a (IC) trade if you can place the short strikes outside a 2 STDEV move and make an average return (incl. losers etc.) of 1% per 5 day period?

A 2 STDEV move happens 5% of the time (per definition) so shouldn't you target 5% plus returns then? (much more actually accounting for the fact that financial markets aren't normal distributed (fat tails, black swans etc.) and for commission, slippage etc) and shouldn't it be independent from holding period as you would adjust your STDEV calculation to your holding period (see question a)

Theortically a two standard deviation move happens 5% of the time -- but that assumes a normal distribution of prices -- which, while the efficient market hypothesis accepts as fact -- is 100% bogus. Any "professor" that tells their MBA students that prices are normally distributed is either (a) a liar, ( an ignoramus, or © uses a flawed model for teaching purposes (most likely answer). I see you note this as well. My entire premise in trading options is that the underlying pricing models are flawed -- whether it be black-scholes, binomial trees, or whatever -- all of those models make assumptions that simply aren't true. That's why its possible to make money on options, if the models were 100% accurate, then only the market makers would profit.

As to my returns, I target a 5% retun on short term ICs -- which, over the past three years, has averaged an actual return of 1.1% per trade. And yes, I'm saying if I stick to my models (which varies instrument by instrument), over the last three years, trading the RUT when I go outside of 2 SDs, holding 5 days or less, using a holding period of seven days and 14 data points when calculating SD AND making sure the trade is equal to or outside of the largest 30 day move, AND making sure I get 5% after accounting for commissions, I average 1.1%. (I assume a mid-point closing price, which isn't always accurate, but I don't know a better way to account for slippage).

However, that trade does not come up that often (haven't gone back and counted, but it's probably about 25% of the time). I also don't make the trade depending on other risk weigting factors. (For instance if I have several longer dated ICs on, or certain calendars -- basically anything where I don't want a market move).

C) I use STDEV a lot looking at the risk and the probability of a return of a trade however not to the extend that you seem to, but I have to say I don't come across too many trades where the market pays you a 10% yield for a trade that needs a 2 STDEV move to make a loss. Even though I have to say I usually look at 1 to 1.5 STDEV especially for shorter dated trades as I don't like selling the small premium of 2 STDEV moves. So maybe the market DOES pay a premium for these however I don't think it's enough over 5% to account for the factors mentioned above. So I don't think you'll beat the market on probability (maybe as a MM receiving the spread rather than paying it) which leads me to question

The trades don't come up that often -- at all -- I run about 50-60 weeklies through my SD model each week. Typically at most I come up with 3-4 possible candidates. (And yes, I account for earnings, past announcements, etc).

D) how many of these 2 STDEV trades have you done over the last 3 years? You say you have been profitable doing them but even if you did 1 of these trades a month (so 36 trades) the chances are pretty good (~16%) that you have been 'just lucky' to avoid a 2 STDEV move. You can be 'profitable' quite a long time selling the 5% chance of a big loss for a 2% premium before the odds catch up with you.

I average 2-3 a month -- so about 80-100 over the past three years. However, for each one I trade, I did significant back testing on (5-7 years worth of data). I would create my rules (so on RUT: (a) strikes have to be outside of the largest move in the last 30 days, ( strikes have to be at or more than a 2 SD seven day move over the last 14 days, © I have to get at least a 5% over margin, after commissions, on the trade). On the backtesting, before there were weeklies, if I wanted to backtest for weeklies, I would just have to check toward the end of each option period (using the last 7 days of monthlies to test the value of weeklies).

Once you have all of the historic data, I found that, for most options, you could construct a set of rules that allowed you to hit certain profit margins. Kim was right -- it's much more common to find profit lines on longer dated options, (so selling premium on RUT 6 weeks in advance) -- but others exist too, and I'll try to extract that wherever I can. On individual stocks, its particularly easier around earnings -- but also increases your risk.

Position sizing doesn't help you with that issue by the way as your expected return is still negative over a large number of trades, so while if will prevent you from blowing up your portfolio it doesn't help you to make money if the expected statistical outcome for the trade is negative.

You're absoltuely right -- if your expected return was negative over a large body of trades, it would be stupid to trade these. But, after backtesting, paper trading, and then RL testing, I believe these are positively weighted returns -- that's why I do them. Of course the risk is bigger -- what if I get hit by 3 15% moves in a row? That would be 3 very large losses, which would turn the series of trades into negative territory. However, that is statistically unlikely.

Hope that clears some thigns up, if not, just ask more.

Marco.

[Kim 7,423]

Chris, thank you very much for the detailed explanation. If it works for you, that'e great. For me, it's a little bit outside my comfort zone.

This reminds me a guy who was running a very successful newsletter since 2000. He was so successful that at some point he started to have issues getting filled on his SPX trades because his subscriber base got too big. One of his trades had a value of $80M US.

His strategy was selling one credit spread every month 3-4 weeks before expiration for 3-4% credit. He had 60 winning months in a row - http://www.wickedprofits.com/options-trading-monthly.html. I had a very intensive discussions with him and argued that the credit that he is getting is not worth the risk and he was just lucky by beating the statistics. Then he had a 27% loss in Feb. 2010 and 51% loss in July 2011. His luck simply ended and those losses erased more than a year of gains. He closed the service few months ago.

I guess what I'm saying is that for me, I found that even with high probability of trades, I still find that I'm not comfortable with this kind of risk/reward even if it has a positive expectancy.

[Davewill 2]

Question on RUT weekly expirations date and settlement values.

As I recall, the trades expire on the friday at COB of the closing week,

but the settlement value can fluctuate through to the following monday depending on the actual close values of ALL the RUT components.

So it is possible for you to think that you have a close based on RUT at 780, but the actual close could be 20 to 50 points different if some catastrophy happens over the weekend.

So the safest strategy is to close all trades before expiration. True?

Is this true for weeklies as well as monthlies?

[Kim 7,423]

The monthly RUT options are settled on Friday AM, I think weeklies are the same. They stop trading on Thursday, so yes, if something happens on Friday, you can be seriously screwed. To make things worse, the settlement value can be very different from the opening value on Friday. I personally NEVER hold through expiration Friday - it's called settlement risk. You can be fairly safe on Thursday and find yourself losing the whole credit on Friday.

[Marco 221]

Thanks Chris for taking the time for the detailed response! I still share Kim's reservations towards the risk reward of the trade but I believe now that you put a lot of effort an thought in these trades and are clearly a bit more than 'just lucky' to have avoided a catastrophic loss with these trades.

I would like to double check if I understood your method of calculating STDEV. Would you mind giving me your 2 STDEV move for a weekly trade on RUT for yesterday (or any date in the past if you only run your sheet weekly or so) so I can check I get to the same number.

Cheers,

Marco.

[Marco 221]

His strategy was selling one credit spread every month 3-4 weeks before expiration for 3-4% credit. He had 60 winning months in a row - http://www.wickedprofits.com/options-trading-monthly.html. I had a very intensive discussions with him and argued that the credit that he is getting is not worth the risk and he was just lucky by beating the statistics. Then he had a 27% loss in Feb. 2010 and 51% loss in July 2011. His luck simply ended and those losses erased more than a year of gains. He closed the service few months ago.

Had a quick look at his performance page. Out of curiosity how did he survive 2008 with selling put spreads? Or did he sell call spreads that year?

There is another guy on seeking alpha (who regularly ranks no. 1 or two in the options category) who loves to sell 0.02-0.05$ credit (!) spreads on weekly options to risk 5$. Don't know where to start to begin with what's wrong with that but he insists it's a great trade and calls everyone ignorant who claims the opposite (quite a few to the credit of SA readers)

Chris I'm not comparing what you are doing to that!

[Kim 7,423]

First, let me just clarify that I agree with Marco - what Chris doing it different.

This guy was selling credit spreads, not ICs, so he was directional. he actually skipped October 2008 (as well as few other months like May 2010). He actually gave me access for few months, so I know that the performance is real and accurate. He was pretty good at predicting where the market won't go, but eventually he was out of luck.

As for SA - you are referring to John Mylant (btw, the reason he is #1 is because he is writing about 2-3 articles every day and the ranking is set by page views). I had some discussion with him a while ago but just gave up. He claims to "know" what SPY will do before the end of the week and structure the trades accordingly... And he is doing the same with AAPL.

Check out those guys - http://www.5percentperweek.com/customer/customerMain.php?section=tradePage&step=viewClosedTrades. Great results for weeks, and then a 94% loss, a 75% loss and few 30-40% losses.

[cwelsh 1,497]

Thanks Chris for taking the time for the detailed response! I still share Kim's reservations towards the risk reward of the trade but I believe now that you put a lot of effort an thought in these trades and are clearly a bit more than 'just lucky' to have avoided a catastrophic loss with these trades.

I would like to double check if I understood your method of calculating STDEV. Would you mind giving me your 2 STDEV move for a weekly trade on RUT for yesterday (or any date in the past if you only run your sheet weekly or so) so I can check I get to the same number.

Cheers,

Marco.

My SD, 7 day (difference in prices over 7 days), using the last 14 days of date was 15.30744 yesterday. (so 2 SD is 30.6)

To clarify: I took the last 15 days of closing prices and calculated the seven day price movement over 1 days (have to have 15 days of data for 14 days of price change).

So column three on the spreadsheet looked like this:

=(A1-A8)

=(A2-A9)

=(A3-A10)

=(A4-A11)

.

.

=(A15-A22)

Then I have:

=STDEV(C1:C14)

So I have a calculation of 7 day moves over the last fourteen days.

[cwelsh 1,497]

Chris, thank you very much for the detailed explanation. If it works for you, that'e great. For me, it's a little bit outside my comfort zone.

This reminds me a guy who was running a very successful newsletter since 2000. He was so successful that at some point he started to have issues getting filled on his SPX trades because his subscriber base got too big. One of his trades had a value of $80M US.

His strategy was selling one credit spread every month 3-4 weeks before expiration for 3-4% credit. He had 60 winning months in a row - http://www.wickedpro...ng-monthly.html. I had a very intensive discussions with him and argued that the credit that he is getting is not worth the risk and he was just lucky by beating the statistics. Then he had a 27% loss in Feb. 2010 and 51% loss in July 2011. His luck simply ended and those losses erased more than a year of gains. He closed the service few months ago.

I guess what I'm saying is that for me, I found that even with high probability of trades, I still find that I'm not comfortable with this kind of risk/reward even if it has a positive expectancy.

Well if he had losses that big I'm sure (i) he was not position sized properly and (ii) did not have a good overall portfolio risk management. That situation could NEVER happen to me, it's a mathmatical impossiblity. (Well I guess the world could end or some other doomsday scenario, but in normal trading, that just won't happen). These close in weekly IC's make up less than 10% of the portfolio and can never make up more than 5% of the risk (offset by trades that move inversely). So let's say everything goes to hell in a handbasket and all of my close in weekly IC trades go 50% against me -- worst case scenario I'm down 5% over all. Not a nice month certainly, but one that you survive from easily. I would never just have "one trade," as you note, that's not smart.

[cwelsh 1,497]

Had a quick look at his performance page. Out of curiosity how did he survive 2008 with selling put spreads? Or did he sell call spreads that year?

There is another guy on seeking alpha (who regularly ranks no. 1 or two in the options category) who loves to sell 0.02-0.05$ credit (!) spreads on weekly options to risk 5$. Don't know where to start to begin with what's wrong with that but he insists it's a great trade and calls everyone ignorant who claims the opposite (quite a few to the credit of SA readers)

Chris I'm not comparing what you are doing to that!

To me that's a guy who doesn't understand risk -- I feel like I at least understand what the risks are and have built a trading strategy that first and foremosts says "What if I am wrong....and I mean REALLY wrong....how much would this hurt me?" If the answer is "not much," then I'm much more comfortable.

As always, if you have questions, just ask away. Thanks for the positive comments

[Kim 7,423]

Well if he had losses that big I'm sure (i) he was not position sized properly and (ii) did not have a good overall portfolio risk management. That situation could NEVER happen to me, it's a mathmatical impossiblity. (Well I guess the world could end or some other doomsday scenario, but in normal trading, that just won't happen). These close in weekly IC's make up less than 10% of the portfolio and can never make up more than 5% of the risk (offset by trades that move inversely). So let's say everything goes to hell in a handbasket and all of my close in weekly IC trades go 50% against me -- worst case scenario I'm down 5% over all. Not a nice month certainly, but one that you survive from easily. I would never just have "one trade," as you note, that's not smart.

You are absolutely right.

He had just one trade per month, and he was actually saying time after time that you should place only small part of your account into it. But I guess when people see 3-4% month after month, they just get a false sense of safety. With auto-trading, I'm sure most of his customers didn't even understand what he was doing. His rule was: if the index touches the short strike - get out. His first loss was when he didn't exit before Friday settlement with RUT about 4-5 points below his short strike. Could exit for 5% loss, ended up with 27% loss (could be much worse). The amazing thing is that some of his subscribers placed a 5 star review on StockGumshoe on Friday only to find out on Sunday that they lost quarter of their account..

[cwelsh 1,497]

With all of the above said -- RUT just moved 11 points, which puts it outside my margin of safety today, so I'm out.

That said, I exited for 0.22, after commissions a 2.7% gain. Not bad for a one day trade.

(I've had several of these one day off trades, where I get in on Monday or Tuesday on the weekly and out the next day when a trade moves outside my realm of safety -- always seem to work out -- but I have not done a backtesting of how well they work, of if it works better than any other strategy, that'll have to go on my to/do list. Basically, I'd be curious if the risk was lower for the same, or better, returns).

[Marco 221]

My SD, 7 day (difference in prices over 7 days), using the last 14 days of date was 15.30744 yesterday. (so 2 SD is 30.6)

To clarify: I took the last 15 days of closing prices and calculated the seven day price movement over 1 days (have to have 15 days of data for 14 days of price change).

So column three on the spreadsheet looked like this:

=(A1-A8)

=(A2-A9)

=(A3-A10)

=(A4-A11)

.

.

=(A15-A22)

Then I have:

=STDEV(C1:C14)

So I have a calculation of 7 day moves over the last fourteen days.

Ah I see what you are doing different now. I would only take fri to fri returns (the time I would hold my weekly) so after 14 days I would only have 2 data points from two weekly returns.

I guess you have a lot more and 'smoother' daily data but I don't think the results will be too different.

[cwelsh 1,497]

Ah I see what you are doing different now. I would only take fri to fri returns (the time I would hold my weekly) so after 14 days I would only have 2 data points from two weekly returns.

I guess you have a lot more and 'smoother' daily data but I don't think the results will be too different.

I used to do that, but didn't have enough data points that matched my trading ranges. Yes I'm using 7 days of data for five day trades -- but that actually helps provide more of a cushion. Glad you understand where my calculations are coming from though.

[Marco 221]

First, let me just clarify that I agree with Marco - what Chris doing it different..

As for SA - you are referring to John Mylant (btw, the reason he is #1 is because he is writing about 2-3 articles every day and the ranking is set by page views). I had some discussion with him a while ago but just gave up. He claims to "know" what SPY will do before the end of the week and structure the trades accordingly... And he is doing the same with AAPL.

I know, one of the good things of SA is also one of its weaknesses everyone can post whatever he wants and talk nonsense without quality check. They should consider changing their ranking from quantity to quality and introduce the possibility to rank people or their articles. John would probably still get quite a number of '5 stars' (he has his fans for some reasons) but there are enough smart people on SA who would give him '1star' only...

I still can't make up my mind whether he's ignorant and really doesn't understand the risks or arrogant thinking he can control them or ruthless in just pushing his newsletter and 1:1 advice to make money from people who don't understand options.

I found comfort in the fact that not many people will be able to put on these 5c credit speads as for a few 100$ credit they would have to post a few 10,000$ margin. I just hope that not some retiree puts his life savings in that stupid trades and gets wiped out.