HotStockMarket › Forums › Stock Market How To › Stock Market Education › Quant question / choosing an option strike price
New Posts  All Forums:Forum Nav:

Quant question / choosing an option strike price

post #1 of 3
Thread Starter 

Suppose I'm looking at an options chain on my screen and I've decided whether or not I want a put or a call, but I haven't decided what strike is best.  Additionally, I have a price target in my head and for whatever reason I think the stock will be priced at X on or before day Y.  I want to decide which strike will give me the largest return if I'm right, but I also want to minimize my loss in case I'm wrong.  

 

Is there a mathematical model for this?  Obviously Black Scholes is involved but there are too many variables that it's difficult to visualize.  And I also know zilch about quantitative finance (although I do have a degree in math so if someone points me in the right direction I should be good).  I can sit there and punch in 800 different scenarios into a calculator but obviously with a rapidly changing price you don't always have time for this.  

 

One thing that might be helpful is if there was a tool that would let you specify all but 1 (or 2) variable(s) of Black Scholes and then show you a 2D (or 3D) graph of the function as your variable varies.  For example, let's say I want to see what happens if I keep # of days to expiration and market price the same, but vary strike price.  It should show some kind of 2D graph that might help me make a decision if I feel confident that the price will be X at some point on a certain day (maybe I feel like an equity is going to test some key support point, for example).  Or perhaps I want to see what happens if I vary both market value and # of days to expiration but leave strike the same.  This might help me decide what would happen with a given option if I'm wrong and I let it ride too long, for example.

 

It seems like somewhere in here there's a numerical integration waiting to happen that I could supply a list of "probabilities" that I think represent how likely the underlying is to reach certain prices on different days during the contract period, and it would just spit out an "expected profit value" for each strike price making it a trivial decision which strike to choose.  

 

Am I making sense here?  Is there any kind of tool that helps with this?  Is there a more practical approach to choosing the best strike price to get in on something at?  (Even if there is a more practical approach, I'm still curious if anyone can enlighten me about a theoretical approach).

post #2 of 3

Dude you're long overdue finding this website haha. Its a blessing:

 

http://www.optionsprofitcalculator.com/

 

Day by day how much the option is worth, with whatever strike price you set. Also keep in mind you can set the limits of the Y axis of the graphs to make it show every 50 cents, or every dollar, etc.

 

 

As for the theoretical approach, if you have X price in mind before Y date. You are right as far as how confusing it is. There is no set answer, but here are some guidelines:

 

If you think your X price will hit in merely a couple days, then almost 100% of the time you should pick the option that is closest to expiration. Remember, with every option you are BUYING time. So obviously the more time you buy, the more needlessly expensive each option will be.

 

Now if you think it will take (about) a week or more to hit your X price, then the option that expires next month will almost 100% of the time be your best choice. The reason for this is because the last 30 days of an option the time decay accelerates a lot. So although you will be paying more for the extra time, most of the value will stick around when you sell it a couple weeks later. As with the closest expiring option, that time value you bought (although originally cheaper) will decay so much that it costs you more in loses waiting to strike than the further out option.

Confusing as hell I know haha, but I hope thats understandable.

 

 

Now for which strike price you should chose is a little simpler. The closer your strike price is to the market price, the more money you will have to initially lay out. The reason for this is because the option only has to move a little bit for you to make money. So bottom line, it is less risk, less reward, and the initial investment will be more expensive.

 

For a strike price further away. Lets say just below your X price. It will be much cheaper compared to the above choice. If your stock makes a quick move over your strike price, it is much much more valuable than if you picked a lower strike price. (speaking in terms of % gain). more risk, more reward.

 

And for the kicker to make things confusing: The further away strike price must be broken quickly in order to get that fat reward. This is what you need optionsprofitcalculator for. After a certain period of time, even if you break through your higher strike price, the lower one actually becomes more valuable. Why? Because eventually the intrinsic value of the lower strike price will outweigh the extrinsic value of the higher strike price.

 

...I think I need to lie down after that

post #3 of 3
Thread Starter 

The calculator looks useful, but it seems to have some unfortunate limitations.  For one thing it only appears to support quarterly calls, unless I'm just missing something.  For example type in SPY 120 puts and it says the Bid/Ask is $5.28 / $5.56 which corresponds to the current Bid/Ask of the options expiring at the end of the month.

 

So that kind of diminishes the usefulness, although it does look pretty good if I'm playing with quarterly options.  

 

Also, the mathematician in me really wants to see an algorithm/formula for computing just a straight up "expected profit factor" given a bunch of inputs about how likely I think the price is to touch certain values on or by certain days during the contract period. In other words, an indicator that takes all the subjectivity / emotion out of the decision.  Sure, I can look at a table like this and decide on my risk/reward tolerance but it's not that easy to process an entire table of numbers like that and come to a reasonable decision in a fast paced environment.  

 

In theory it seems like you could just perform a double integral over the rows and columns of that table (preferably over the actual black-scholes though since it's continuous), weighting each price by some kind of "likelihood function" or probability distribution that is itself a function of time and then it would just spit out the exact strike price that gives the highest expected return.  So I'd still be interested if anyone has suggestions for references that talk about this kind of thing.

New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Stock Market Education
HotStockMarket › Forums › Stock Market How To › Stock Market Education › Quant question / choosing an option strike price