This post is a response to Twitter buddy @demonetizedblog
Let me take a stab at a “process” answer.
Introduction
For directional trading 90% of the work happens upstream of the option expression.
The option trade construction is the most trivial part of the process. Your fundamental work should inform your opinion of the distribution. This can be compared with the implied distribution from the vol surface.
This mental process is entirely different from vol trading. Remember, you aren’t dynamically hedging. Directional trading and vol trading have totally different starting points.
[At the end of the post you’ll see when the two approaches come to the same conclusion and when they don’t. This can lead to directional traders to trade with vol traders and everyone is happy. It’s still zerosum. It’s just that the losses can be incurred by whoever provided the liquidity to the dynamic hedger. That entity was not part of the original trade]
Ok, so when it comes to directional trading vs vol trading, you must be clear what game you are playing.
This post is about structuring directional trades.
What’s the distribution?
First, you do a bunch of fundamental voodoo and come up with a distribution of possible stock returns.
[I’ll wait]
Good. We are going to discuss options now. Relax. Take a breath. Don’t worry about fancy words like “moments of a distribution” or kurtosis. You are a fundamental investor. It’s fine to think in prices, percentages, and bets.
Now what?
Let’s establish a focusing principle.
You want the short leg of an options spread to correspond the most likely landing spot of the stock based on your analysis. If those options are the cheapest on the board you might want to consider that the option surface is not presenting you an opportunity. It agrees with you. Don’t rush over that. This is not intuitive. Many fundamental managers buy the strike of where they think the stock is going. Don’t do that. Instead let’s review some basics about distributions. Without real math.
 A biotech stock worth $100 might be trading for that price because it’s 90% to be 0 and 10% to be $1000. True bimodal.
Codeswitching this idea into options:
 The 100 call is worth $90.
 All the OTM 100 point wide call spreads are worth $10.
 All the butterflies are zero.
What are some courses of action here?
Let’s say you can afford 1 100 strike call. You could have also chosen 9 900/1000 call spreads. Or 3 of the 700 calls. In this case, all the propositions are the same because the options are correctly priced.
[Prove this to yourself. I’ll wait.]
Cool. Now you can imagine how if some of the options were priced differently you might be able to find an alluring proposition.
 New stock to consider. An insurance company also trading $100. This is not a bimodal stock. Perhaps it looks more like a bell curve with a high peak shifted to the right of the forward price because a pumped up put skew is signaling strongly negative skew.
Wait. Why does that push the peak to the right?
Think about it. For that stock to be $100 with a long left tail, it must have a greater than 50% probability of going up. The verticals will show you that. It’s the opposite case of the biotech stock and with much less volatility.
 If you were super bullish you might want to load up on the depressed slightly OTM calls.
 If you were bearish but thought the left tail was not as long you might want to buy the .50d/.25 put spread to express the view by exploiting the excess skew you think the market is embedding in the OTM puts.
Just remember, options give a shape to the distribution. Not every $100 stock has the same distribution. Think about where the $100 comes from? What upside force is counterbalancing the downside? The biotech stock has a very long right tail 900% away counterbalancing a large mass of probability that’s only 100% away. The $100 stock price is nothing like the insurance company. Options allow you to express the bet you want to express. The stock price alone is too blunt.
Once you let that simmer you can start to ask yourself useful questions:
 Would you rather own 1 atm call or more calls for a total of the same premium at a higher strike?
 Now compare that to call spread candidates. How many call spreads can you buy and at what moneyness?
The nice thing about vertical spreads is they cancel out many of the “greeks” effectively taming your vega and gamma exposures. The bets can be thought of as binaries allowing you to make simple over/under bets. To calibrate your impression of the possible magnitude of a stock move, you consider the moneyness or how far away from stock price the chosen strikes are. The moneyness will depend on your intuition for the volatility of the stock. You will have a sense for which spreads are “close” or “far”. These are technical terms.
And since I mentioned volatility, let’s say a few words on that to help you avoid some landmines.
Is the vol cheap or expensive?
If you are a directional trader you don’t care if the right volatility for an option is 55% or 56%. You aren’t dynamically hedging. But you don’t want to go to the usedcar lot without at least checking Carmax online. You can compare the implied vol to the distribution of historical realized to make yourself feel like you did diligence.
Here’s a simple way:
Compare the IV to the stock’s historical vol of a comparable tenor. So if you are considering a 6 month option look at the distribution of 6 month historical vols to see if you are on the high or low side of the range. How? Looking at a vol cone will get you a quick optical answer.
Here’s Colin Bennett’s example (with my highlight) from his book Trading Volatility:
If the recent realized volatility is elevated and you wanted to buy longdated options it might be a poor time to buy options. You can either wait, trade structures like verticals that have little vega exposure, or even create a directional trade by selling options.
Here’s a few extras to consider when selecting an expiry:

 The nearer the option tenor, the more event pricing matters. The event’s variance is a larger proportion of the total variance until expiration.
 Longer dated options have takeover risk. (Cash takeovers mean your LEAP extrinsic goes to zero. Sorry.)
 Do you plan to roll the exposure to maintain it or is there an expiration to your thesis? The more often you roll the less rebalance timing risk. This has to be weighed against trading costs.
The Real Work Is Not In The Options
When you throw a proper punch the fist is just the delivery method. The point of contact. That’s the option expression. The real work happens from the torque in your hips. That’s the fundamental analysis behind the punch. An advantage of directional trading is you can think in discrete bets once you’ve done your fundamental homework.
Discrete trades let you:
 Think in terms of how many bets you get paid back vs how much premium you layout and compare that to the probability your fundamental work suggests.
 You’d like to get to a statement that looks like “I’m willing to risk 1 bet to make 3 because I think the proposition is a 50/50 shot.”
 This establishes your expectancy and shape of the p/l.
 Combine that info with your bankroll and now you can size the trade.
Bonus Section: Volatility Traders
I said that directional trading and volatility trading are different games. I’ll briefly talk about that.
First of all, even vol managers sometimes make discrete bets. They will “risk budget” a trade. I’m willing to spend $1mm on 150% calls for winter gas. Or whatever, you get the idea. They might even set up a separate account for tracking and attribution for this.
But really this risk budgeting or discrete framework is different from managing a relative value volatility or market making portfolio. In that environment, you are often responding to values moving around some crosssectional trading model. You see edge, you pick it up, throw it on the pile and manage the blob. With a decent size book holding thousands of line items you are going to need 3D goggles to slice and dice the positioning and the risk. You might not even know what you are rooting for sometimes. If you are short SPX correlation and long 200 of the 500 names then you are massively overweight vol in the 200 and you are “synthetically short” vol via the index in the other 300. Hundreds of names x hundreds of strike x hundreds of expiry and you need to bucket and compute quickly and accurately. Totally different animal from directional perspectives.
This does not mean that vol traders and directional traders don’t land on the same conclusions occasionally. A vol manager who finds a name that “screens cheap” might be looking at the same thing a fundamental manager is seeing. The fundamental manager is coming from a different vantage point, but might feel that a stock is hiding some serious upside and the nominal price of the calls are a bargain. In this case, the fundamental manager is going to struggle to find liquidity as the call options might be cheap for a few contracts but once they start calling around the street find that no market maker is willing to join the resting retail offers.
You may be wondering why the screens are so low in the first place? Why are they stale? The market maker’s dashboards are flashing green too. They know those options are cheap. But remember this is a game. They aren’t going to bother lifting the offers for a few contracts. They would rather freeroll on the possibility that some donkey overwriter who systematically sells calls without price sensitivity dangles a mid market offer. Then they’ll lift. (gratuitous “Do You Even Lift Bro?” clip)
So when do vol managers and directional traders trade with each other? All the time. Here’s 2 examples.
 Imagine a fundamental trader who is directionally smart but not vol savvy. They might buy calls, and the market makers who have been keeping tabs on this pattern of flow realize its predictive of a price move but has not historically beaten them to implied vol (perhaps it’s one of these dumb accounts that buys the strike of where they think the stock is going. They should probably hire a vol trader, if for nothing else to show them how to do p/l attribution). So the market makers sell the calls and overhedge the delta. Trading 101.
 A very common case where directional traders and vol traders are happy to trade is on vertical spreads or ratio spreads. Say a directional hedger buys put spreads. Vol traders can be happy to sell them so they can buy that tail option that the hedger gave them as the lower leg of the spread. A similar example would be a 1 x 2 ratio put spread. Say the stock is $100 and the directional trader buys the 80/75 1 x 2 put spread for a cheap or even zero premium. In their mind, they make make money all the way down to $70. They don’t start to lose money until the stock has dropped more than 30%. The vol trader has a different view. The vol trader cares about path and they know if the stock trades down to $80 quickly and vol explodes, they are going to be long vega and have ammunition to sell into the panicky vol buying. That 1 x 2 put spread is going to mark ruthlessly in the directional traders face. The directional trader didn’t respect path. Option traders are extra wary of path because they are highly leveraged businesses warehousing complex portfolios with nonlinearities. There’s no better training for visualizing risk up, down, through time, across correlations, and at different speeds. The trader who honors path will often be the reason that “option that will never hit” is priced so high.
If you liked this post, consider checking out the Moontower Volatility Wiki.
Hi
I would like to thank you for the very interesting content you share in your posts.
very explanatory for those people like me who want to learn to use options as something more than an instrument for directional bet. thanks .
It’s also amazing to hear stories about market makers. uou I don’t undesrtand how they do it . Hundreds of names x hundreds of strike x hundreds of expiry ,
A simple and prob dumb question , for a vol trader, ultimately everything comes down to the difference between realized and implied volatility?
It is easier to have a forecast on vol that a forecast on direction ? it’s because the vol its mean reverting ?
It has always intrigued me how MM decides at what iv they will price their options.
thank you
There’s many forms of edge. IV vs RV trading is one flavor of trading. Vol is meanreverting but everyone else knows that too. So as you might expect the market consensus IV reflects that. If rv is very elevated than you will find a descending term structure for example
Sorry am I missing something, shouldn’t the 1000 butterfly for example have some value because of the bimodal distribution? In this case 10c?
Perhaps a silly question, but shouldn’t the 1000 butterfly have a nonzero value? Say 10c?
Yes, the 900/1000/1100 fly is worth $10
I should have been more clear that all the equidistant flys up to 1000 as the top leg are equal in value. Thanks