Anyone here trade options?

[FIRE'd@51]

Are you guys saying that Options House charges a total commission of $10 regardless of whether you sell 10 naked puts or 10 credit spreads?

[utrecht]

Quote:

Originally Posted by FIRE'd@51 Are you guys saying that Options House charges a total commission of $10 regardless of whether you sell 10 naked puts or 10 credit spreads?

They have 2 commission structures and you can switch back and forth between the two.

1) Smaller traders.
$5 for up to 5 contracts. $1 per contract above 5
$8.50 per spread up to 10 total contracts. $1 per contract after 10 contract

2) Larger traders
$8.50 per option plus 0.15 per contracts
$12.50 per spread plus 0.15 per contract

So it would be $10 for 10 naked puts and $15.50 for 10 spreads.

[utrecht]

Quote:

Originally Posted by cb7010 first..I'll apologize for your losing a few hours of sleep..then..I will disclaim that I am pretty much a beginner with options and frankly have not been able to take as much time as I'd like to work through all of the math and study the back data as you have. And, thus, my original questions and desire to learn more. I am probably doing myself a great disservice by not factoring in margin to my calculations, prefering to take the easy way and base my calculations on 100% cash backing. I am trying to figure out a way to acheive income, with managed risk and without a ton of adjusting and worrying down to the wire on expirations days. Thus, I am always sell OTM credit spreads, and specifically the condors (bull put spread and bear call spread, both OTM) and balancing the return with my comfort level and VIX at the time of the trade. For example, if I look at the SPY right now, 125.48, and at next Friday's 111111 options..in theory before the bell yesterday, I could have done the 120/119 (.07 credit) on the Put side and the 131/132 (.06 credit) on the Call side..my total risk is $100 per contract, as with any "condor" only one side can go bad, so total at risk is only one side of the spread; and, my total GROSS credit is $.13...a couple side notes..the SPY might not be best suited for these types of spreads, I've been doing mine on individual shares, and as you and a previous poster point out, you must use OptionsHouse or your transaction costs will eat up your profit; but, in the example above..you have a roughly a 4.4% cushion on both sides and a gross 13% return (again, not factoring in margin or commissions). So, unless my math is way off some where, this is intriguing to me, in fact, I'd be happy with a 1% per week return if I could broaden the condor out further for more safety. I am still working on figuring out the best ways to accept losses and/or roll out when one of the spread are in danger and rationalizing if there is an inherent problem in my logic because of option volumes and commissions; but, again, I am intrigued by the possibility and I love glancing at the market and not really caring if it's going up or down most of the time knowing that my spread is wide enough to accept most normal flucutations. I really just need more time with my calculator and data..in the meantime, I've just been doing a few spreads and condors to get my feet wet and haven't been burned yet..

The math on these things gets confusing but how is your total risk $100 per contract? Isn't your total risk $87 per contract? The most you can lose is the difference between the strikes minus the credit you took in. Since only one leg can be a loser, you risk $100 minus your $13 credit.

Also calling that a 13% return (which would now be 14.9%) is a bit misleading. I say misleading because, like I mentioned in an earlier post, returns on options cant be compared easily. Someone might see a 13% or 14.9% return for a week and think this is a get rich quick scheme. I guess its fine as long as you're only using it to compare to other possible option trades that you are considering.

[FIRE'd@51]

Quote:

Originally Posted by utrecht The math on these things gets confusing but how is your total risk $100 per contract? Isn't your total risk $87 per contract? The most you can lose is the difference between the strikes minus the credit you took in. Since only one leg can be a loser, you risk $100 minus your $13 credit. Also calling that a 13% return (which would now be 14.9%) is a bit misleading. I say misleading because, like I mentioned in an earlier post, returns on options cant be compared easily. Someone might see a 13% or 14.9% return for a week and think this is a get rich quick scheme. I guess its fine as long as you're only using it to compare to other possible option trades that you are considering.

I would say that returns can be compared easily. Return is simply the $ made or lost divided by $ invested, which is the average margin required to carry the position until it is closed. It is the risk that is more difficult to compare because of the non-linear payoff pattern of options.

[utrecht]

Let me see if I get better explain what I meant by "returns on options cant be compared easily". I just showed a few posts ago, that selling ATM naked SPY options is fairly comparable on a risk / reward basis to selling 3% OTM put spreads. The amount of potential profit compared to the amount of margin required to put on those trades is pretty close.

However, if you try to compare the returns they aren't close. Over the past 13 weeks, the average credit for a 5 strike 3% OTM SPY spread is 0.58. So max profit is $58 and max loss is $442 for a max profit of 13.1%.

The average premium collected when selling an ATM put is 2.08 with an average SPY price of 119.02. Whats your max return percentage on this trade? Do you divide the $208 into your max loss like you did on the spread trade? You cant because its unlimited.

When I figure returns on my naked put trades, I divide the profit into 100*SPY because that's the amount of money it would cost to have those shares put to me even though I never let them get put to me. Technically, 100 shares could also get put to you in a spread if you dont close it out so you could use that same formula for percentage return on a spread trade, but that's not fair either. It will always be much lower because the premium collected is always much lower.

I just dont see how you can look at a naked put trade and a put credit spread and try to compare them apples to apples using a percentage of return.

The only way I see to do it is to divide max profit for each trade into your required margin for each trade. However, when you do this, your return is nowhere near 13% which is why I said that number is misleading and very confusing when used for comparison to any other kind of trade.

[ERD50]

Quote:

Originally Posted by utrecht Let me see if I get better explain what I meant by "returns on options cant be compared easily". ... However, if you try to compare the returns they aren't close. ... I just dont see how you can look at a naked put trade and a put credit spread and try to compare them apples to apples using a percentage of return. The only way I see to do it is to divide max profit for each trade into your required margin for each trade. However, when you do this, your return is nowhere near 13% which is why I said that number is misleading and very confusing when used for comparison to any other kind of trade.

You can compare returns just as FIRE'd@51 described. There is nothing misleading and/or confusing about it.

What is harder (as he says) is to compare the risk.

For the spread, since the max you have 'tied up' is that $5 spread (minus the credit) that is your denominator. So ~ 13% in your example.

For the cash-covered put, assuming you need the entire cost to buy the put, your denominator is $119 minus the credit. Yes, the % return is very different, ~ 2%. It is what it is.

Now look at risk - with the spread, it is true you only have ~ $5 'at risk', but there is a very real chance that you can lose ALL of it. There is very, very little chance (cats and dogs living together...) that SPY will go to ZERO and you lose it all. So the risk/reward profiles are different. But overall, the total risk/reward is very similar when you average it all out (ot investors would flock to the one with the lowest risk/reward, and supply/demand would bring it back in line). Another way to say that - you get a higher max-gain/max-loss ratio with the spread, but a higher chance of experiencing that max-loss. With the put, a much lower max-gain/max-loss ratio, but a much lower chance of experiencing anything near that max-loss.

-ERD50

[utrecht]

Everything you just said is true, but I would say that most people who aren't experienced option traders would find this confusing.

You just a compared a trade with a 2% return to a trade with a 13% return and I can easily show that they will have total returns that are very close over the long term. Explaining that to the next 100 people I come across and having more than 5 of them understand it would not be easy.

[ERD50]

Quote:

Originally Posted by utrecht Everything you just said is true, but I would say that most people who aren't experienced option traders would find this confusing. You just a compared a trade with a 2% return to a trade with a 13% return and I can easily show that they will have total returns that are very close over the long term. Explaining that to the next 100 people I come across and having more than 5 of them understand it would not be easy.

But would they understand this (lifted/adapted from this thread):

Roulette analogy:

Player A, on a SINGLE NUMBER - the payout is 35:1, but he will only win 1 of 38 times.

Player B) On RED - the payout is 1:1, but he wins almost half the time (18/38).


Yet, the house will, on average, over the long run, collect the exact same 5.263% of the money bet from each of them. There is no difference long term, despite the huge difference in payouts for a win. Options are approximately the same game. You can change your payout, but not have a huge effect on your long term return.

-ERD50

[utrecht]

I get your point but its not exactly the same. In Roulette you either win or lose. That's pretty easy to understand, even if you don't know the exact odds. With options, there is a lot of room in between the max profit and max loss.

[clifp]

Quote:

Originally Posted by utrecht I get your point but its not exactly the same. In Roulette you either win or lose. That's pretty easy to understand, even if you don't know the exact odds. With options, there is a lot of room in between the max profit and max loss.

Well you could make a complicated analogy with two slot machines with 96% payoff. E.g. One machine which makes a lot of payoffs where you get your bet back, and the other one which only pays off $1,000 or more.

Of course the problem with these analogies is makes option trading sound a lot like a gambling. When of course we all know that option trading really is sophisticated form of investing, designed to reduce portfolio volatility

Ok that is what I tell myself

[FIRE'd@51]

In order to get an idea of the expected excess return, I integrated the short put payoff pattern weighted by a log-normal stock price distribution with an annual mean of 7% (approximately the historical growth of the S&P 500 price) for a one-week at-the-money put and came up with an expected weekly return of 0.063% (3.27% annually) for a cash-secured put, before commissions. This means that if you were long the S&P 500 and wrote at-the-money puts using your long position to fully collateralize the puts, you would expect to add 3.27% annually to the return of your index fund. I believe this is the way that utrecht is calculating his returns.

I used a 30% annual volatility (about where the VIX is today) for both the implied and realized volatilities, effectively assuming the puts are priced efficiently. The result is pretty much insensitive to this number so long as the the realized volatility equals the implied volatility.

The nice thing about using weekly options is that you get many "draws from the urn", so I would expect one to get pretty close to the expected return over a couple of years of following this strategy. On the other hand, commissions will greatly cut into the return. From what has been posted above, it seems that the commission at Options House is about 0.01 per share. On an SPY of 125, that is 0.008% per week, or 0.42% per year, which would reduce the net excess return to 2.85%. And that's not counting the times you might have to close out the position to avoid assignment, thus paying another commission.

Of course, this is an expected return. Many weeks the returns will be considerably in excess of this number, as utrecht has thus far experienced. My number takes into account the possibility of a several standard deviation event, such as was experienced in 1987. It is the finite probability of such a "black swan" event that drives down the expected return.

I invite any of the mathematicians here to check my results.

[My Dream]

Sorry to sound ignorant but I'm not sure what trading options are. I did a google search and did some reading and still got somewhat confused. Does anyone know of a site that can explain it in simple terms so I can understand?

[utrecht]

I'll admit that I dont fully understand all of your math. Having said that, I have a few things to throw at you.

1) There was a study done a while back (still available on the net somewhere) that showed that selling covered calls against your SPY holdings gave you an increased return with lowere volatility. IIRC the study showed your returns were 2-3% higher annually. Of course selling puts is basically the same thing with some slight differences. That study was done before weekiles were available. My research shows that the weeklies increase the return a significant amount above the monthlies. My point is that loing term I believe that my returns over and above whatever SPY returns will be significantly more than 3.27%. Although Ive only been doing this in real time for about 6 months now, during those 6 months the market has seen several months of average returns, one month that was one of the worst month in history and one that was one of the best in history. I dont think I couldve had a better 6 months in which to see how this strategy works. My returns during this time are MUCH better than what your math shows. So far (off the top of my head) my returns are closer to 14% higher than SPY. I dont mean 114% of SPY returns. I mean SPY is down like 3% and naked puts are up 11%. (Quoting from memory..dont have my spreadsheet with me).

2) As far as commissions, obviously the more contracts you are trading the less affect commisions have. When trading 10 contracts like I am doing currently, it does work out to about 0.42% in commission costs.

3) I dont pay another commission when I have to close out the position to avoid assignment. It just turns into a spread. I buy to close the expiring option and sell to open the new one for next week. One commission. It does cost a bit more depending on how many contracts are being traded but its not much.

[FIRE'd@51]

Are you referring to the BXM study? IIRC, the implied volatilities of the calls sold were systematically higher than the subsequent realized index volatility over the time period of the study, which pretty much explained the outperformance.

Having said that, I do think implied volatilities tend to, on average, be higher than the subsequent realized volatility, due to market makers fear of gap moves (such as in 1987). In my calculation, I assumed the implied volatility from the put price equaled the subsequent realized volatility. Mathematically, I can calculate the effect of this by changing the standard deviation of the probability distribution in the integration in my spreadsheet. If I price the puts using a 30% volatility and then assume the realized volatility is 29%, the annual excess return (before commissions) jumps to 6.15%. If the realized volatility is 28%, it rises to 9.03%. This could be what was going on in the period you were doing your trades. Obviously, if the options you sell are systematically overpriced, that only enhances the performance of the strategy.

[utrecht]

Quote:

Originally Posted by My Dream Sorry to sound ignorant but I'm not sure what trading options are. I did a google search and did some reading and still got somewhat confused. Does anyone know of a site that can explain it in simple terms so I can understand?

I believe I mentioned a few posts ago that everything having to do with options, especially calculating returns is very confusing.

Maybe try this?
Stock Options For Dummies, free PDF download

[FIRE'd@51]

Well, I went back and calculated the weekly percentage price changes of SPY from 5/13 - 10/28, when you were doing your real time trades. The annualized standard deviation over those 24 weeks was about 25%. According to the Black-Scholes model, an at-the-money put with a 25% volatility would be priced at about 1.4%. In post # 67 you said your average put price over that time period was 1.77%, and you said your puts were slightly out-of-the-money. This is the effect I was talking about in post # 94.

Although the market has been considerably more volatile than usual over the past 6 months, it appears that put options were priced in anticipation of even higher volatility, and evidently you (as a seller of volatility) benefited from that.

[utrecht]

Are you saying that the only reason that I greatly outperformed the returns for SPY so far is because of put options being overpriced due to high volatility? If so, I believe you are wrong.

Basically all I am doing is selling time. The shorter the time frame until expiration, the higher the percentage of put premium is time value.

Ive back tested this going back to the beginning of 2005. There was plenty of time during that 7 year stretch when volatility was not higher than normal and selling naked SPY puts still greatly beat a B&H strategy of the Sp500.

The back test is approx due to the fact that historical option prices are not available anywhere, but its pretty easy to approximate using today's options prices and adjusting for whatever the VIX was each week.

From 1/1/05 to 11/11/11, these naked weekly puts have a CAGR of 17.2% vs a CAGR for SPY of 2.5%. There's no way that can be explained away. The fact is that most of the value of an option within a week of expiration is time value and it melts away quickly.

Your last sentence says that I benefited from selling volatility. I'm actually benefiting from selling time. I'm not yet sure that I will make more when volatility is high. I will collect more in premiums each week but I also stand a higher chance of taking a big hit. Its like one insurance company selling insurance only to teenage drivers and one only to 40 year olds. The 40 year olds pay less money in premiums but they also crash less so the insurance company keeps the premiums a higher percentage of the time. Which scenario makes more money for the insurance company long term? I'm guessing it works out about the same, but one thing is certain. The insurance company always makes money over the long run.

[clifp]

Quote:

Originally Posted by utrecht Your last sentence says that I benefited from selling volatility. I'm actually benefiting from selling time. I'm not yet sure that I will make more when volatility is high. I will collect more in premiums each week but I also stand a higher chance of taking a big hit. Its like one insurance company selling insurance only to teenage drivers and one only to 40 year olds. The 40 year olds pay less money in premiums but they also crash less so the insurance company keeps the premiums a higher percentage of the time. Which scenario makes more money for the insurance company long term? I'm guessing it works out about the same, but one thing is certain. The insurance company always makes money over the long run.

I think you and myself (although I've made the mistake of selling longer term options on the believe that volatility would die down) have made good money on the perceived/implied volatility on a daily/weekly basis when there actually has been almost no movement over the last three months.

Here is some interesting data from the M* Dividend Investor weekly email.

Quote:

I can understand a certain amount of apprehension on the part of investors; I feel it too. I'd hate to think that policy mistakes in Greece or Italy will hurt the American economy; let's face it, we've made more than enough mistakes of our own. What I don't understand is why so many investors are chasing their tails. Stay invested (which is my preference) or get out, but don't try to time this stuff. Here's one way to ascertain what the markets have been dishing out through this mess so far. As of Friday's close, the Dow Jones Industrial Average closed at 12,153.68. That represents a gain of 10.44 points since the end of July. But in the 74 trading sessions on the way to posting that negligible gain of 0.09%, the Dow gained 6,781.52 points on up days and lost 6,771.08 points on down days. All told, that's 13,552.60 points worth of price changes, more than the entire value of the index just over three months ago--and also more than double the sum of daily price changes in the same-sized period that preceded it. So much sound and fury, so little to show for it.

That is average of 1.5% change per day in the value of 30 of Americans largest companies. Yet the fundamentals, revenues, profits of these companies probably have not changed by more the 1.5%/quarter.

To use your analogy, we are getting to sell expensive insurance to 20 years old, but because of the the round the clock gridlock (picture roads in Chinese megacities), the 20 years can't drive more than 20 MPH and aren't getting hurt. For all the drama, with debt crisis in the US and Europe we have over the last 3 months, fundamentally the economies haven't really changed.

[utrecht]

Agreed, but this is not something that only works in this environment. People pay too much money (by buying puts) for insurance to protect their portfolios against a decline.

Weekly naked puts have outperformed each and every year going back to at least 2005 which is where I stopped, including making money in 2008 when the SP500 was down 37%.

[FIRE'd@51]

Quote:

Originally Posted by utrecht Are you saying that the only reason that I greatly outperformed the returns for SPY so far is because of put options being overpriced due to high volatility?

No, I'm not saying that at all. You outperformed SPY because SPY was down 4.6% over the past six months that you have been implementing your strategy real time. Of course you will outperform SPY in a down market when you are taking in put premiums every week. As you said, selling cash-secured puts is identical to following a covered-call strategy, and we know a covered-call strategy outperforms when the underlying asset goes down.

What I said was that I believe your outperformance was enhanced because the puts you sold were overpriced relative to the subsequent realized volatility.

If we have a strongly rising market for several months, I would expect your strategy to underperform SPY. No surprise there. That's what happened from September 2010 until February 2011. The expected return I calculated in post #91 is a statistical result based upon an infinite sample size over all types of markets.

Quote:

Originally Posted by utrecht Basically all I am doing is selling time. The shorter the time frame until expiration, the higher the percentage of put premium is time value..................... Your last sentence says that I benefited from selling volatility. I'm actually benefiting from selling time. I'm not yet sure that I will make more when volatility is high.

Don't be fooled by the relatively higher put premiums per unit time obtained by selling weekly puts. That is to be expected and follows simply from the fact that volatility per unit time increases with the square root of time. That is to say, an annualized volatility of 32% is equal to a daily volatility of 2%. That is why a one-month put cost about twice as much as a one-week put, not four times as much. If both weekly puts and monthly puts are fairly-priced, you would get the same results following a monthly strategy as a weekly strategy. As I said above, a weekly strategy gives you a lot more times to play the game, so you should reach a statistically significant number of trials sooner.