ThinkOrSwim's Probability OTM, ITM, of Touching Calculator

C

cactiman

1. How does this calculator come up with the % number? I have a basic understanding of Historical and Implied Volatility, Standard Deviation, the Greeks, and the Black Scholes model. Is it a combination of all these things?
2. How accurate has anyone found the thinkorswim % numbers to be? Anyone back tested it?

Thanks in advance for any comments or suggestions.

:)
 
The percent is based on standard deviations. 1 SD = 68%, 2 = 95%, etc. The amount of data used for the computation depends on how many days until expiration.

Accuracy? Well it's freshman level math, so I gotta go with 100%. Reliability is another issue.
 
I don't like how I put that. The computation is always based on one years worth of data for the SD. And then the percent is adjusted for how many days are left. I think you knew what I meant. Just wanted to be clear, though.
 
It makes a bunch of assumptions about how prices will move in the future based on how they moved in the past. Then it uses those assumptions to guess.

It's wrong, a lot.

It's also "right" (in the Price is Right sense) a lot.

Use at your own risk.
 
Quote from jnbadger:

I don't like how I put that. The computation is always based on one years worth of data for the SD. And then the percent is adjusted for how many days are left. I think you knew what I meant. Just wanted to be clear, though.
More...


Extra thanks! (on top of my "in advance" thanks)
Would you care to make any comments on the "reliability issue"? You sounded skeptical.
By back-testing the "accuracy" I was asking more about how often trades based on them have succeeded, etc. (i.e. reliability)
Some people base their trades on these probability % numbers, but the numbers change all the time...
Are they meant more for a rough guide perhaps, just a reconfirmation the "odds" are in your favor?
:)
 
I agree with random capital. Just use it as a general guide. I know you can use the think back feature, and I think you can punch in a date which will give you percentages for the upcoming expiration. But be careful. Something about backtesting options strategies scares me. I'm pretty good with options knowledge, but you need someone who's better at it to chime in at this point.

You're welcome. Happy to help. That's why we're here. Well, most of us.
 
Quote from Random.Capital:

It makes a bunch of assumptions about how prices will move in the future based on how they moved in the past. Then it uses those assumptions to guess.

It's wrong, a lot.

It's also "right" (in the Price is Right sense) a lot.

Use at your own risk.
More...


I see. Thanks for your thoughts.
Not very good at my "thanks in advance" bit am I!!
:D
 
Quote from jnbadger:

I agree with random capital. Just use it as a general guide. I know you can use the think back feature, and I think you can punch in a date which will give you percentages for the upcoming expiration. But be careful. Something about backtesting options strategies scares me. I'm pretty good with options knowledge, but you need someone who's better at it to chime in at this point.

You're welcome. Happy to help. That's why we're here. Well, most of us.
More...

You and Random.Capital are actually reconfirming my own doubts about "trading by the numbers". I still think an "organic" reading of the Stock/ETF chart is the way to go.
For instance, if there was a big crash 6 months ago, for geo-political reasons or some other extreme event, those numbers (deviations) are averaged in. But now things might be calm and the Stock/ETF is just going sideways in a small range, so those big numbers don't apply anymore, etc.
:cool:
 
I was just discussing the same issue on another board.
I consider it a "lazy and risky" substitute for doing ones own analysis.
Here is what I wrote:

<<< I've been puzzlling over your 5 pt. spreads with 95%
probability outcomes. >>>


Personally, I think statistical "probability outcomes" are somewhat meaningless.
Such outcomes should be based more on common sense and thoughtful analysis, than a generic math formula.
While the formula may be helpful in getting some vague general fluctuating idea, at that particular moment in time, common sense should have already given you a general idea of the likelihood of success.
If not, you probably should not be doing the trade.
Whether the numbers calculate out to 77% or 93% chance of success is less relevant than my own analysis.

For example, assume all variables are the same on a 6 week naked put or credit spread, including 2 stocks that are both 13% otm.
But one is trading no where near any kind of tech support, and the other is once again trading at it's multi tested level of tech support, per the 1 - 2 year chart.
Given a limited bankroll, which trade would you rather initiate.
I'd make the decision based on analysis and common sense, and pick the one at multi tested tech support,... not a coin toss of statistical probabily, based on where the stock happens to be trading at that moment in time.

I'd also base it on a few basic fundamental criteria pertaining to the companies financial health. While such issues may not be very relevant in the world of option trading, I look at those criteria to evaluate whether a stock has the "potential to recover", if a bad market takes it down.
I'm less inclined to panic sell, if i know I have not invested my cash in an over valued, and/or DEBT LOADED piece of crap..... regardless of how limited the loss may be.

% probability outcome formula's, can give investors a false sense of confidence.
Nor do they speak to the likelihood of the stock recovering in a reasonable period of time, if a bad market suddenly takes it down.
Initiating a trade based on a generic statistical outcome formula, is a "lazy and risky" substitute for thoughtful analysis and common sense.

Those 95% probability outcome formulas really are meaningless.
Better to go with 96%.
<< g>>
 
Quote from cactiman:

You and Random.Capital are actually reconfirming my own doubts about "trading by the numbers". I still think an "organic" reading of the Stock/ETF chart is the way to go.
For instance, if there was a big crash 6 months ago, for geo-political reasons or some other extreme event, those numbers (deviations) are averaged in. But now things might be calm and the Stock/ETF is just going sideways in a small range, so those big numbers don't apply anymore, etc.
:cool:
More...

The calculations are based on implied volatility so what happened 6 months ago doesn't matter. Implied volatility reflects the view of option market makers of what might happen to the stock before said option expiry... Of course they can be wrong + it is well documented that extreme 2 or 3 standard deviation moves are underpriced because reality is poorly reflected by normal distribution.
 
You must log in or register to reply here.
Back
Top