I entered the Bear Put Spread in QQQ at $0.37 at market open yesterday. This puts the max risk of the trade at $0.37 and the max gain at $0.63. The ETF sold off throughout the day leaving the position up $0.05 at the end of day one. I’m leaving the trade on for now.
While looking at my daily charts today I noticed that the Nasdaq is looking at a possible break down. This could be a good opportunity for a short term bear put spread.
Chart of the QQQ on May 23rd
The idea is to bet on continued downward trend over the next day or so. To implement the strategy you can buy the in the money June 57 put and sell the June 56 put. Your max profit is then $59 dollars per contract and your max loss is $41 dollars per contract. The following chart shows a risk analysis of the trade:
Risk Profile for the Trade
When I first started investing I was a strict advocate of fundamental analysis. I read “Security Analysis” and tabbed multiple sections with post-its and wrote notes throughout on the margins. I still go back and reread sections of the book frequently. I am still a believer that fundamental analysis is valuable. However, last year I read a book by Mandelbrot called “The (Mis)Behavior of Markets”. In the book he shows evidence that trending does in fact occur in financial markets.
If trending does occur is it possible to profit from trends? There are a number of traders who have done so. The most well-known in the popular press are the “Turtle Traders” although others have also developed systems to trade based on trends.
To test whether trends do exist I downloaded the closing price for the SPY from 1993 to Aug 2010. This obviously is not a full scale test as I looked at one security over a relatively short period of time. It is possible that trending occurs at different strengths in different markets. I actually think you may find more trending in more volatile markets but I would have to pull a lot more data to see.
Number of days in trend was defined as consecutive days up or down. Probability is defined as the probability that such a trend would happen given an even chance of the security moving up or down in a given day. Observed is what actually happened. Delta is the difference between the two.
This is obviously a very simple analysis. It is interesting that the trend continues for two days at a greater frequency than would be expected and then the market seems to over-correct in the other direction.
Of course, most trend following systems do not simply buy any position for a day based on the idea that they will move up more than they move down. Many also hold the position for greater timeframes and build in allowances for daily variation within some limit. In this case, what if we say that instead of just looking for day over day consistent movement in the same direction we look to see if 3 days in the same direction is an indicator that over the next week the price is moving higher.
In order to get an idea of what this longer-term trend would look like I first calculate how many times during this period the SPY closed higher/lower 5 trading days after a 3 day run-up/sell-off. What I found surprised me. There were 347 times where the price moved three days in one direction or the other. In these cases, the SPY closed in the same direction 5 trading days later, despite some drawbacks in between, sixty percent of the time. The average move was only about 1% but the range was a full 10% meaning that there could be opportunity for occasional big returns.
This was obviously a simplistic way of looking at the market and I would not suggest simply opening a position any time a stock has trended in one direction for 3 days. It does indicate however that some trend following elements can be built into a successful trading system.
Work unfortunately has continued to be crazy meaning that I did very little trading over the last month. I am seeing a light at the end of the tunnel with the day job though and have already been able to open more positions in the last week than I have in the last two months.
My loss ratio on short options positions has continued to trend down. I show zero for August and September only because these positions have not yet been closed:
Overall, I have not caught back up to the S&P 500 performance for the year yet:
This marks the one year point from when I started this experiment. Most of my losses came from two long only stock positions I had. During the last year I executed 135 short option positions and a handful of long option positions. I will be posting an analysis of those trades once I get some time. Overall I am down 5% for the year.
One of the things I love about investing with stock options is the ability to make a bet on the market so many different ways in order to take the risk you want and minimize cost. I was thinking about this a few days ago while looking at options on MSFT. For the purposes of this exercise I looked at the quotes from end of day Friday. The following table shows different ways of opening a bear call spread on Microsoft:
I have written before about how I believe that trading options where the underlying is close to the strike is generally better. This chart may show that although I need to think about it some more. It seems like the sell price of the far out of the money strikes is too low. This probably has to do with the standard pricing models being based on a normal distribution but I will need to dig further to see if that is what is going on. If that is the case you could make the argument for buying far out of the money options. I’m not claiming that now but I have filed it away as something to look into in a future post.
The thing to notice here is that the spread with the two closest strikes is the one with the best risk/reward ratio. As the second spread moves further out of the money the ratio gets worse and worse. In the case above though we are looking at a short option position. What happens if we reverse this and look at a bull put spread? (Note added 15 June: this should read “bull call spread”. Essentially we are looking at the two sides of the same trade, selling/buying a call spread)
This table seems to confirm the need to dig deeper. In this case the risk/reward ratio improves the wider the spread. Of course,I have not looked at probability of expiring in the money for either of these examples but it does seem to suggest that when selling short options go for narrow spreads and when going long look for wide spreads.
There are a number of things I will look into further:
1. How does probability of expiring in the money play into this?
2. Are the far out of the money options priced as suggested by the standard models and if so is there an opportunity to profit from non-normal price movements?
3. Does the risk/reward skew between long and short positions mean that the short options game is for suckers? What is the comparable expectation values?