On Friday, March 15, I moved my APR 900 butterflies to the APR 930's. The difference in prices of the two flies was $10.11. To make matters worse, I'd already had to move those flies once from their original center strike as the RUT continued its relentless northward climb. As the RUT repeatedly hit and moved past the upside breakeven, the long-delta positions (a deep-in-the-money call and some call debit spreads) somewhat cushioned the position, protecting it from devastating losses. Those long-delta positions didn't prevent all unrealized losses, of course.

However, combine the cost of the protection and moving the flies, and I was nearly up to my pre-planned maximum margin in the trade. Future adjustments would have to reduce or maintain the margin and would have to be made with that limit in mind.

That limit for me is $5,000/butterfly contract. On the day the trade was entered, the original investment was only $3,117.86/butterfly contract. If I hadn't set aside monies for these adjustments, I might have been forced to close out a possibly viable trade and lock in the small unrealized loss. Alternately, I might have been forced to roll into fewer contracts of butterflies than in the original position in order to reduce the total cost of rolling. I compute my margin/butterfly contract based on the original number of butterflies on the day I entered, because that $5,000/butterfly contract maximum for that month was computed based on that number. Worse, if I had a different temperament than I do or was less experienced, I might have been tempted to forgo a needed but costly investment because "it's gotta turn around."

How did I know ahead of time that $5,000/contract might be a workable amount to set aside? I back-tested the butterflies without regard to how much I was spending, basing my decisions only on the methodology I was using. I looked at the maximum margin that was employed during the period that I was back-testing, throwing out the outliers on the high side. I found that almost all months, somewhere between $4,000-5,000/contract appeared to be a workable amount to spend on my preferred strategy and all subsequent adjustments.

What if you don't have access to a back-testing platform? I certainly couldn't afford to pay for the more expensive platform when I first began trading. One method I used was to paper trade through at least several cycles. Then I started small with the live trades until I could determine what I might typically spend. I certainly didn't want to size up and then find out that the trade and its adjustments resulted in more margin--and, therefore, risk--than felt comfortable for me. I also spent some time with the tools that the platform I was using then provided. These were somewhat clunky theoretical options pricing tools and a rudimentary analysis chart.

For example, imagine that you trade iron condors and your plan is to roll the call or put debit spreads out of the way if the absolute value of the delta of the sold call or put reaches 16. You plan to roll into enough new spreads to make up the loss you took when you sold the original spreads. Let's see what we can work through using Think or Swim's Analysis Page. We'll work with the put side of that iron condor, so instead of putting the whole condor up, I'll just use the put credit spread portion. For the May cycle, I've chosen the following for the put side of the iron condor:

Sell May 13 1420 for $4.95
Buy May 13 1410 for $4.45.

When I was trading iron condors for many years, I preferred to find the first put with a delta above -0.09 (-0.08, for example), but this put had a delta of -0.09. It was the first 10-point spread that returned at least $0.50 credit, however. The low volatility at the time required selling closer to the money in order to make the minimum amount of money to warrant the risk.

Since I'm basing my adjustment on the 1420, I would typically put that on a risk graph and run the graph forward in time while dropping the price and increasing the implied volatility. The implied volatility generally increases when price drops. The risk graph you'll see below comes from Think or Swim, but I could do this even on OptionsXpress, with its less flexible risk graphs.

For example, I might want to look at the possibility that the SPX could drop to retest the 2/19/13 peak high by March 27. Remember that I rough my articles out and snap charts a week or two before the articles are published, so charts will not be up to date.

Theoretical Analysis Graph of May 1420 Put on 3/27, with IV +4.00 Percent:

Under those conditions and this adjustment plan, this trade would need to be adjusted. The sold put has reached the delta level that was mentioned in this imagined adjustment plan.

If I were trying to estimate the cost of this adjustment, made under the same conditions as set up on the chart above, the May 13 1380 put would have a delta of -8.92, so it would be an appropriate choice for the sold strike of the new 10-point credit spread. Not all readers use Think-or-Swim, so let's calculate the cost of moving the flies from a tool available to all: the CBOE's free Options Calculator.

Sample Theoretical Calculation for the May 13 1420 Put on March 27 with SPX at 1532 and IV Rolled up 4 Percent:

The days to expiration, volatility, and price of the underlying have all been set up to mimic the same situation seen in the previous graph. The theoretical value of that put, using these inputs, was $10.73. The same exercise returns the values of the other puts. Using this theoretical price of the 1420 put and the other puts, we can calculate that it would cost $10.73 - 9.00 = $1.73 (at least) to buy back the in-trouble credit spreads. A credit of $5.09 - 4.14 = $0.95 would theoretically be available for the 1380/1370 credit spread. The trader might also look at the 1370/1360 credit spread, which would bring in a credit of $4.14 - 3.34 = $0.80. Depending on my market view, I might prefer the $0.80 credit for the ten extra points of safety.

In accordance with a plan used by many iron condor traders--although I have some quibbles with it--the intention is to sell enough new credit spreads to make up for the loss on the other spreads. That loss is $1.73 paid to buy it in - $0.50 original credit = $1.23 plus commissions. To make up the loss, I would have to sell 1.23/0.80 = 1.54 new credit spreads for each old one I closed. For example, if I'd originally had 2 contracts of that iron condor, I'd want to sell 3-4 new put credit spreads to make up either most or all of the difference.

That makes up the loss and allows for the same profit on the iron condor, but it increases the margin. Hang in there with me while we calculate the increase. Remember that we're utilizing the clunkiest--but free and available to anyone--tools to imagine what an adjustment might cost. This is to show that even those without backtesting tools and with only the clunkiest tools can estimate how much adjustments might cost before the trade is entered.

Imagine that we'd originally pocketed the same $0.50 for the call credit spread portion of the iron condor. The original margin, then, was ($10.00 spread between contracts - $1.00 credit) x 100 multiplier x 2 contracts = $1,800 plus commissions. Remember that margin is held for only one side in a balanced iron condor. Unbalance it by having 10-point spreads on one side and 5-point spreads on the other, and you may be paying margin for both sides, depending on the brokerage.

The new margin = $10.00 spread x 100 x 3 contracts - ($0.80 credit for new put spread) x 100 x 3 contracts - $1.00 credit for original contract x 100 x 2 contracts + $1.73 loss x 100 x 2 contracts = $3,000 - $240 - $200 + $346 = $2906 plus commissions. Therefore, considering that the trader started out with a two-contract position with a margin of $1,800 or $900/unit, that trader needs to set aside at least $2906/2 original contracts = $1453/contract plus commissions if this type of adjustment plan is anticipated.

Admittedly, this is a clunky way of testing the cost of adjusting in the future and determining how much money might be needed for adjustments. It's so clunky that I probably made some simple math mistakes in those calculations: I'm far better at the calculations involved in higher-level math than I am at the simpler ones. However, I wanted to demonstrate what you can do with even clunky tools to plan the maximum margin you want to invest in a trade. You can then determine if a certain adjustment plan interests you or is viable for your situation and account size.

It's dangerous to open a trade thinking you'll adjust it without allowing enough money for the adjustments you had hoped to make. Through the years, I've heard from traders who had to close viable trades at a loss because they didn't set aside enough margin or buying power to make adjustments.

Be aware that I've used the iron condor because it avoids talking about the tricky way some butterflies are margined after adjustments. If you know your adjustment plan for butterflies, it's a good idea to talk to someone at your platform about the way that specific platform would deal with the margin or buying-power effect with the adjustments you plan.

I also want to add a caveat to the iron condor adjustment plan this article discussed. I chose that plan to discuss because it's relatively easy to calculate the cost of the adjustments, even with clunky tools, and because it's a common adjustment plan used by many iron condor traders. However, it can get traders into trouble, too, and may not be workable in all conditions. I adjusted that way for several years, with success. However, about three years ago, I had two months' worth of trades on at the same time. I had plenty of money set aside to roll into double positions one day when the trades ended up near the adjustment point but not quite there yet. My underlyings had been running up but were all near strong resistance, so I wasn't tempted to adjust early. As it turned out, it wouldn't have mattered, even if I had adjusted early. The next trading day, the markets gapped up and ran hugely. In a runaway market, it was time to roll into double positions on two months worth of trades. Because the implied volatilities were so low, though, I could roll up only about 10 points, if I wanted to collect any premium at all for my work. Doubling all the positions would have meant I would have more than $200,000 at risk in a runaway upside market.

Did I want to risk more than $200,000 to move the call credit spreads only ten more points in what had already proven to be a runaway market? No way, I decided. That, however, forced me to lock in the unrealized loss and close out the trades, with that now realized loss much larger than my planned maximum loss. That trading methodology may not put enough emphasis on watching unrealized losses since the intention is always to roll into more contracts. The lesson here is two-fold. First, whatever your adjustment plan, always pay attention to unrealized losses because they may have to be realized. Rethink your trade or your adjustment plan if it lets unrealized losses grow too large. Second, the impossible happens sometimes. You may not be willing to put your adjustment plan into effect due to something happening in the market or in your own life. Always watch what your "impossible to happen" maximum loss will be and think about that, too, when you plan the trades that interest you.