Last week, we started a discussion on risk. In particular, we discussed a delta-neutral strategy, a short stock position combined with a bull call debit spread on AAPL composed of the following strikes: +5 660 07-SEP-2012 12/-5 670 07-SEP-2012. This position was theoretically entered on Friday, August, 31, and the chart below shows the strategy summary of the position at initiation.

Strategy Summary of Call Spreads:

At initiation, the position was delta-neutral. Since delta tells (or, rather, suggests) how much an option position's value might change with each one-point change in the underlying, we know that over a small price change, the position is relatively immune to losses. That will change if price moves accelerate. We used this theoretical position last week to validate a point Jim Bittman makes in located in a section titled "Establishing Risk Limits" in Trading Options as a Professional: setting up a delta-neutral trade doesn't mean that the trade is without risk.

If we've understood that setting up a delta-neutral position doesn't mean that we've set up a risk-free position, we then have to decide how much risk we want to take on in a trade. We're often accustomed to setting that risk at a certain dollar amount or percentage amount we're willing to lose, basing that amount on a percentage of monies we have invested in the trade. Then we look at a price chart and estimate where the underlying's price might go. We might do some quick multiplying of likely price change by the delta to figure out how much our profits or losses could be at the targets we've identified on our charts. If we're good with Greeks, we'll know to take a look at the gamma, too, and use that to compute how delta would change over that likely price change.

If we're visual people, we might look at a risk analysis chart like the one we could have viewed at trade initiation and shown in last week's article. AAPL has moved since that Aug 31 theoretical trade initiation and the options involved have expired, so this graph is not currently accurate, of course.

Strategy Graph

If we understand a little something about volatility, we might roll volatility down a little if we want to see how profit or loss would be impacted if prices were to rise. We might roll implied volatilities up a little or a lot to see how profit or loss would be impacted if the underlying's price falls.

But maybe we're going at it backwards, Bittman might suggest. Maybe we should start with the vega and likely volatility changes rather than price and likely price changes. Vega measures how much a position's profit or loss will change with a change in implied volatility of the underlying. Bittman specifically suggests that you figure out how much a typical change in volatility is for your underlying, look at the vega for the trade, and multiply the two to figure out whether you can accept the risk in the trade.

For example, imagine an OCT iron condor AAPL trade. I don't think I would have put on an AAPL iron condor right before that September new-product release unless I thought I could bear a lot of possible financial pain between the trade initiation and that release. As it turned out, I checked the profit/loss chart for the trade late Friday, September 14, and the trade was down only $90.00, with price pretty well centered in the middle of the iron condor. The chart below depicts the strategy summary for a position theoretically set up on August 31, employing regular monthly cycle September options. It was a 10-contract SEP iron condor composed of the following contracts: +10 SEP 730 Calls/-10 SEP 720 calls/-10 SEP 610 calls/+10 SEP 600 calls.

Strategy Summary for Iron Condor:

Vega is -122.69. For each 1 percent change in implied volatilities, this position's profit or loss will change by approximately $122.69, if all other parameters stay the same. Since vega is negative, the position profit will theoretically drop by that much if implied volatilities rise by a point and will rise by that much if implied volatilities drop by a point. In other words, if AAPL's price sits still but implied volatilities had risen into the new-product release, the position might suffer a loss, although that loss would be somewhat offset by the passage of time.

What's a typical variation in implied volatilities? If you know your underlying well, you'll have some idea of how it and its implied volatilities behave. If not, you may have to look for other sources for the information.

The following OptionsOracle chart suggests that at 45 days to expiration, which would be the next trading day after this theoretical position was established, a standard deviation of AAPL's volatility was 4.51 percent. If a trader considered that a one-standard-deviation move was a reasonable guess for the volatility, that means that the trader risked a 122.69 x 4.51 = $553.33 profit or loss due to normal, run-of-the-mill changes in the implied volatilities alone. If the trader thought that implied volatilities might move two standard deviations, due to the upcoming product release, the risk was double. A $553.33 risk from changes in implied volatility represented 6.13 percent of the total margin withheld in the trade, after commissions of $1.25 per contract. A loss double that represents more than the maximum possible gain in the trade.

Historical Volatility versus Business Days to Expiration:

Later in the trade, the absolute value of the vega shrinks, so even though implied volatilities might change more, they're being multiplied against shrinking vega values. In the beginning, however, the iron condor trade is as much about changes in implied volatility as it is about changes in price. This is depicted by the relatively large absolute value of the vega when compared to the delta. You might be thinking to yourself, sure, but if the price sits there while implied volatilities skyrocket, what does it matter, since I'm not going to be closing out the trade with the price right in the middle of the expiration profit-and-loss shape? I can just ignore that theoretical and as-yet unrealized loss. That's dangerous thinking.

What if the price doesn't just sit there? What if it drops precipitously, so that adverse price movement also contributes to additional loss? The unrealized loss could grow well past a workable level of loss for the trade. If you're ignoring theoretical, unrealized losses along the way--as I once did with iron condors--you're taking on far more risk than you think you are. I know this from experience.

Bittman went further in his explanations. When traders are considering delta-neutral trades with a stock component, he suggested, they should focus on either gamma or vega risks. There's no need to focus on both. For delta-neutral positions without a stock component, the trader has to make a decision. '"Do I hope to profit from time decay at the risk of losing from rising implied volatility or from a big stock-price change?" or "Do I hope to profit from rising implied volatility or from a big stock-price change at the risk of losing from time decay?"'

Are you willing to accept the risk reflected by likely changes in volatility? If not, the trade you're considering might not be the trade for you.