Quandary: The Logic of Trading Iron Condors

This post originally appeared at The Options Zone


I'm an iron condor (sell call spread and put spread with
same expiration date
on same
underlying asset
) trader.  I understand the risk and reward potential for these positions and recognize that skillful risk management is crucial to my success when trading these negative gamma positions.

The best part of trading iron condors occurs when the markets are calm and no adjustments have to be be made.  In those rare situations, I eventually cover my shorts and bank the profits. 

In the current volatile market, there is little peace.  Depending on the strike prices chosen, iron condor traders are making frequent adjustments to their positions.  That thought took me down a strange path.  Looking ahead by a few days, it occurred to me:

  • It's very likely the underlying asset will move sufficiently so that the iron condor is no longer delta neutral (due to negative gamma)
  • If that happens, the position will have lost an unknown number of dollars

That leads to this quandary:  Why should I trade the iron condor today, when it is very likely to trade at a higher price soon?  Why not simply wait for that higher price to enter the trade?

If I were to follow that unusual line of thought, I would proceed as follows:

1) Choose a specific iron condor and estimate the premium I can collect now

2) Do not enter the order.

3) Wait until that trade is losing money: i.e., I can collect a higher premium 

How much higher?  How long to wait?  I don't have answers to those questions.

4) Open the trade and collect that higher premium.

Obviously this trade is no longer delta neutral, nor is it likely to be the trade i would select at the future date.  But it's the position I would normally open today.  By waiting I get to make the same trade at a better price.  Doesn't that have to be better than owning it at a worse price?

By the time I enter the trade, the position may require a minor adjustment. No problem.  Open the slightly adjusted position in place of the original. 

To adopt this idea, I cannot allow much time to pass or I may miss a profitable trade opportunity.

Thus, the quandary:  Why trade now?

5) By collecting that better price: 

I have more risk than I prefer when making a new trade, but to compensate, I collect additional premium.  The bottom line is: I would be facing this extra risk had I already owned the position, so why not own it several days later – with the same problem, but at a better price?

Another point to ponder:  Is this truly taking on more risk?  By not opening the position sooner, I had zero risk of loss (or profit) during the time that I had no position.  The easiest method for avoiding a disaster is to be out of the market.   Is the benefit of a few days on the sidelines enough to offset a bit of extra risk when the non-neutral position is opened late?   I believe it is.

6) And if that quandary is not sufficient, why not take an additional step?  Instead of waiting, open an unbalanced iron condor right now.

I would be forced to make the trade with a choice of extra upside or downside risk, rather than allowing the market to force that decision upon me. 

I can go even farther by opening half the trade with an upside bias and half with a downside bias.

The real question is:  does it make sense to open an iron condor position at the time I am ready to trade, or is it better to either:

  • Wait for a better price
  • Open two unbalanced IC now, in place of a single, neutral IC


Opening positions that are not neutral when lacking a market opinion feels wrong.  Yet, the probability is very high that I'm going to have that position anyway. And soon.

Why not take advantage of that by opening those two non-neutral iron condors at a better price now?  (I do that by choosing an iron condor that is already unbalanced.  Because it is no longer neutral suggests that it is losing money.)

Bottom line:  It appears that I can take advantage of the fact that whichever iron condor I choose is likely to be a (hopefully temporary) loser at some point in time.  By waiting for it to become a loser and delay initiating the position until that time, a better premium is collected.

Logic tells me this is not sound reasoning.  Yet, I cannot find the flaw(s) in my argument.   A quandary.



, , , ,

28 Responses to Quandary: The Logic of Trading Iron Condors

  1. Jason 05/31/2010 at 11:13 AM #

    Hi Mark,
    If the market drifts higher, or even if it “melts up”, I’d expect implied volatility to drop (as is its come off quite a lot in the past couple of days). I stick with opening delta neutral to slightly delta negative IC’s (to compensate somewhat for the increase in volatility if markets go down), and my general feeling is that when I open one, its more a view on trying to capture that consist implied/realized premium. If you think the underlying is going to move around a lot and threaten your IC, why not just go for either selling a condor or a strangle and then rolling it into a gamma – position when you feel the time is right?

  2. Rajesh 05/31/2010 at 11:26 AM #

    Hi Mark,
    I am dubious about my own credentials as to whether I can comment on this. What seems to me to be a reasonable approach could be to wait for a move of say 1 (substitute x, depending on comfort) SD from some reference point and then open a condor say at Y SD from that point. Would this lead to less adjustments?

  3. David 05/31/2010 at 12:32 PM #

    My two-cents is that some time ago I back-tested exactly this on SPX since 1990 and while I can’t remember what criteria I used for finally entering the position, it always came out with a poorer result in the end over just entering as originally planned.

  4. Mark Wolfinger 05/31/2010 at 12:42 PM #

    This is why it’s a problem. I open positions as you do, but note that there is almost NEVER a month when the IC declines day after day until I decide to exit.
    At some point it is losing money. Why not enter at that time?
    It’s not that I ‘think’ anything specific about the future. It’s history tells me there is going to be a better time to open whichever IC I open today.
    Whether to try to take advantage of that historical ‘fact’ – that’s the quandary.
    Thanks for the contribution.

  5. Mark Wolfinger 05/31/2010 at 12:47 PM #

    I’m sure you are qualified as anyone else to venture an opinion on this.
    If I do as you suggest, I would still be opening the ‘correct’ iron condor after the move occurs. I’m wondering whether I can get away with opening the originally selected iron condor choice – at a later date.
    I believe your suggestion would lead to fewer adjustments for two basic reasons:
    a) less time remains
    b) The position is neutral to start – and when you have no market opinion, that’s the best place to be.

  6. Mark Wolfinger 05/31/2010 at 12:50 PM #

    Glad you looked into this.
    However, I’m not sure which thesis you tested.
    ‘Waiting’ vs. ‘opening now, but non neutral.’
    I don’t see how it can be the ‘waiting’ choice. By definition that must lead to a better result every time one trades. The downside is missing some profitable trades when the position never produces a proper entry point. Thus, even though every trade is better, the overall profit can be less.

  7. Jason 05/31/2010 at 1:08 PM #

    My personal experience is that when I try to “time it”, I usually wiff it and lose money. I generally do a lot better throwing on the position and adjusting as needed. I’m fairly comfortable with my ability to handle positions in the heat of the moment, but all the numbers I’ve been keeping suggest to me that I have absolutely no ability to time IC’s. The IC trades are for a personal account, I trade them because they require relatively less maintenance, I have fun doing it, and I’m not too terrible at it. Unlike my day job, I don’t really have the same sort of pressure to grab every penny off the proverbial table, so I’m happy leaving it like that. Am I optimizing my Sharpe ratio and all that? Probably not, but that’s OK for me.
    Why not just keep a tab of it, either via paper trades or something else to see if you’re right? Might be a long term experiment, but it might be fun.

  8. Jason 05/31/2010 at 1:34 PM #

    Sounds like there may be 3 different groups then:
    1) Delta Neutral IC – the control
    2) Delta Biased IC – Let’s set this at say +/- 5 deltas
    3) Waiting IC – This is the tricky one, but you probably have to put the most parameters on this or it gets dangerously close to back-trading.
    So you run paper trades with this, all with the same tenor, but with no adjustments (in theory adjustments should add value, and I think what we’re trying to isolate is what group performs best).

  9. Rajesh 05/31/2010 at 1:50 PM #

    You are right in that the ‘right’ condor is opened ‘after’ a move has been made. But statistically speaking it should be less likely that another significant move would be made in the same direction. Of course, markets have their own mind and just to spite one would happily march through the condor too! Again it could also go through the short strike in the reverse direction! But I think the overall probability of adjustments over a long period may be lower.
    Let me know your thoughts on this. I must congratulate you on a wonderful blog! So many insights and as you rightly reiterate many a time, a basis on which to dwell on ideas that you have put forth and modify and adjust to one’s own requirements.

  10. Mark Wolfinger 05/31/2010 at 2:56 PM #

    Thanks or the comments. Yes, my goal is to get readers to think for themselves, providing points of view that they may not have previously considered.
    I disagree with your premise. When you flip a coin, the probability of its landing on heads is unaffected by previous tosses. Similarly, an up move should not preclude another.
    But I know how you feel. After that x SD up move, it’s temping to fade the move. But I don’t believe there’s any statistical advantage.
    Unless you have evidence that suggests consolidation occurs more often than statistically predicted after an x SD move, to me the future ought to be just as unpredictable as the past. That means (to me) an equal chance to significantly move in the same direction.

  11. Mark Wolfinger 05/31/2010 at 3:00 PM #

    It’s that ability to ‘handle’ positions that provides your edge. That’s the reason you can earn a profit.
    I too recognize that I can be more effective as a trader with an unknown quantity of additional work. I am not willing to devote the time and energy required.
    One point: I cannot be ‘right.’ I merely mentioned this is an unanswered question. I believe it will remain unanswered. It would take far too many months of trading to provide a statistically relevant number of data points. I leave that to someone who is much younger.
    Yes it may be fun. But I doubt I’ll be trading in 10 years – and even that is only 120 data points.

  12. Mark Wolfinger 05/31/2010 at 3:04 PM #

    The study can be made.
    Omitting adjustments does not ‘feel right’ – because each position is different, it may turn out that the adjustments make so much difference that the original purpose for running the test could be over-shadowed.
    Agree. Which performs best provides good information. Lacking adjustments, what I cannot be certain is whether it provides enough data to adjust one’s approach to IC trading.

  13. Joe 05/31/2010 at 7:37 PM #

    Very interesting quandry and one that is difficult if not impossible to come to a conslusive answer without testing.
    Funnily enough I have been thinking along these lines a week or so back and started opening positions in small increments at different prices as the market moved each day. So now I own the same ICs that were opened at 3 different prices on different days. I have no real evidence as yet (way too early) as to which ones are peforming or will perform better, but can say I feel comortable owing these ICs with differeent short deltas.
    As a side note: In the environment that we are currently in, I find owning unbalanced ICs provides some cushioning and by placing the call side ‘closer to the money’ allows for an upward move/IV drop to make more sense than traditional even ICs.
    Maek, I don’t know how you or others manage their porfolio of ICs but for myself, focusing on my porfolios delta is an effective way of analyzing and assessing which positions need to be opened or closed.

  14. dave appel 05/31/2010 at 8:35 PM #

    Hi Mark,
    You could scale into the whole condor. Lets say you trade 50 condors. you could open it at 10- then go to another 10-20- and then another 10- 20 an so on. This approach allows not only for better pricing but the ability to not have to adjust so quickly because essentially the position size “is your stop” or risk manager. It takes advantage of the markets mean reveting tendency . With the volatility I am now more a fan of trading at reduced position size than making frequent adjustments. I have adopted this methodology and it is more profitable.

  15. Burt 05/31/2010 at 10:22 PM #

    Mark, I think the flaw in your argument, if there is any, might be between static and dynamic estimates. When you put on an IC, you expect the underlying to remain within a certain range. Yet the underlying is always changing. So, if you do wait, do you still believe the underlying will remain in the same range as your original estimate? Has the underlying moved so much that the original estimate is still valid or not? Furthermore, I believe the fundamental thesis behind IC’s is mean reversion — the underlying may move close to the short strikes, but the thesis is that it should eventually back-off to capture the near maximum profit. Thus, if you wait to capture a better price at the same strikes, you’re implicitly saying that you’ve got even more confidence that the underlying will back-off. But this is at odds with the idea that you’re not trying to forecast the market. Hence, the paradox: the thesis relies on mean reversion, but as the market moves further, believing in mean reversion might change into a bet on a significant directional reversal.

  16. Mark Wolfinger 06/01/2010 at 7:40 AM #

    1) Scaling into the trade is an intelligent way to proceed. Open some, open more – if the price improves, and repeat once more.
    If you own the same IC at three different prices, then by definition, the one with the best price will perform best – assuming you manage risk of the entire position as a unit. There is no reason to manage them differently.
    2)Delta is by far the most obvious, and simplest way to manage immediate risk.
    The beauty of options is the ability to use the Greeks to measure other risk factors. Then the trader can eliminate, reduce, or ignore any of those factors. You choose to concentrate on delta. I would never tell you that is ‘wrong.’ I feel the need to modify gamma when adjusting, but that does not suggest that you should do so.
    3) Very true. The put side gets punished by the market far more than the call side, all things being equal. But that one-year market rise was large enough to punish the call spreads – despite a continuous drop in IV.

  17. Mark Wolfinger 06/01/2010 at 7:47 AM #

    Hi Dave,
    Agree scaling is a very good approach and is what Joe was suggesting (previous comment)
    The idea of actively using position size as a risk management tool is a good one. I already do that – but the idea of combining it with scaling works well. Because you intend to increase size – if the opportunity presents itself – it’s unlikely you would be feeling the need to adjust the original position. Any such adjustment would be included when the next trade (at better prices and modified) is added to the portfolio.
    Adding new trades, at better prices, is already a risk reducer. And you can slightly modify that new trade to include a delta (or other Greek) adjustment so that it’s not only a better price, but it’s also safer at the time the trade is made.
    Good idea. Thanks for sharing.

  18. Mark Wolfinger 06/01/2010 at 7:54 AM #

    Interesting point.
    I’m not certain I depend on mean reversion. When the underlying moves enough, I become sufficiently afraid of additional loss so that I do not hold the position as it was. I ‘fix’ it.
    Thus, I don’t truly depend on mean reversion. And, prefer not to wager on directional reversion.
    Yet, you make a very good point. Do traders who ‘hold’ and don’t adjust depend on mean reversion, or are they just hoping for the best? I surely have no reply.
    Burt, maybe you can help me here: ‘Mean reversion’ – does the time frame matter?
    In other words. when placing an IC, am I automatically wagering on mean reversion beginning right now – at the time of the trade? What about the recent past? Shouldn’t trading patterns over that time be considered? (I don’t consider them). Perhaps the stock is reverting to a different mean – one that is far different from that being used by the iron condor trader.

  19. John 06/01/2010 at 10:01 AM #

    Mark, I think IC will generate loss when it is far from expiry. No matter which way the underlying price goes, and regardless of the volatility accompanying it, the magnitude of the winning side is always less than the magnitude of the losing side. That puts effective timing as something irrelevant to IC. This also explains why it is a good idea to initiate an IC using options that are going to expire soon. I agree with you that bigger gamma is bad, but I also believe that’s part of the deal. On the side note, I think holding through expiry may not be as bad as long as the strikes are not too near to the underlying price.

  20. Mark Wolfinger 06/01/2010 at 10:36 AM #

    There are many appropriate methods for opening, holding and exiting trades.
    You and I are on opposite sides. And neither of us is wrong.
    1) I am not comfortable holding trades to expiration. You are. I am not comfortable opening front-month trades when expiration is nigh. You are.
    2) No one suggested that the winning side gains more than the losing side loses. But it can happen when implied volatility levels shrink and time has passed.
    3) I don’t accept your reasoning. ‘Bigger gamma’ is not part of the deal, unless you choose to make it part of the deal. I choose less gamma. You choose more gamma because you want more theta. that is your choice.
    4) Your last statement sounds good, but it is the reasoning of someone who has not seen what can happen. Holding to the end to collect a small remaining premium is a very bad idea – but that’s my opinion. I do not have the data to back it up.
    The recent past (is Oct/Nov 2008 so long ago that you cannot remember?) should tell you that unexpected events occur far more often that one would anticipate. The market was moving up or down by 5% frequently. It’s amazing how quickly people forget.
    I’m not telling you it will happen again this month or this year. But it will happen again.
    5) You should trade what is comfortable for you. If it’s front-month, then it’s front month.

  21. Burt 06/02/2010 at 8:07 AM #

    Mark, you make an important point about the time frame of mean reversion. I would guess that an IC trader would implicitly believe that the underlying will be within a range during the duration of the trade or trending less than the short strikes. So the IC trader doesn’t have to be wedded to mean reversion strictly. Just be right about timing. Paradoxically, it would seem, that although the IC trader supposedly has theta in his favor, he still has to forecast the time of the position accurately.
    As you point out, you’re prepared to fix the position if the underlying moves too much. Does the adjustment reflect the belief that mean reversion will continue, that one was wrong about the appropriate mean, or that the market will slow down, as you note?
    I guess that I may have emphasized the mean reversion thesis too strongly.

  22. Mark Wolfinger 06/02/2010 at 8:34 AM #

    Hi Burt,
    I don’t know about others, but I cannot afford to open positions based on my expectations for market direction.
    Thus, my adjustments reflects none of the beliefs you mentioned. It’s simply recognition that the market has moved too far for comfort and that risk has reached, or is approaching, a point of discomfort.
    I cannot speak for others, but I never know if mean reversion will continue, whether I chose the wrong mean. I trade iron condors when the risk/reward meets my needs. I cannot rely on market expectations. For those who can, more power to them.

  23. David 06/03/2010 at 6:58 AM #

    Hi Mark,
    Yes sorry. That wasn’t clear. I was actually referring to the ‘waiting’ choice. Where waiting was defined along the lines of
    – Decide what position to take and at what time.
    – Wait until position price is better by some factor and then open the position.
    I can’t recall what, if any, adjustment strategy I applied when testing. I may see if I get time to re-run such a test again. The simplest would be to apply the above criteria for varying factors from say 1.0 – 2.5 and make no adjustments to see which comes out best. (Not suggesting that would be an actual strategy to follow).

  24. Mark Wolfinger 06/03/2010 at 12:12 PM #

    We are not on the same page at all.
    Waiting to decide WHICH POSITION is standard fare and is not remotely remotely to the quandary.
    The problem I posed dictates choosing the position NOW and then deciding when to open it.

  25. Burt 06/03/2010 at 12:24 PM #

    Mark, I’m bewildered by your reply. I’ve been reading your blog for a while, so I think I understand your admonition to consider one’s comfort level when trading. Nevertheless, I find it hard to believe that you don’t have some sort of underlying assumption if not about market direction, at least about it’s relative range when you make a trade. I understand adjusting before the underlying gets to a level that reaches your discomfort point. But short of exiting the position, when you roll up/down or out, it would seem that you’ve made an implicit bet on market direction. That is, you expect the underlying to reverse before it touches the further out strike. Maybe you don’t look at it that way. I could imagine that you only look at the risk/reward of the new position without forecasting market direction. But it would seem to me that the risk/reward of the new position depends on the probabilities of the underlying’s movement, which still implies some sort of bias. Even if you don’t consider the probablities, the premium you collect reflect’s the market’s view of the probabilities, so you’re implicitly betting against the market since you’re selling premium.
    This isn’t a criticism of your method. But as I think more and more about your suggestions I find it hard to to conceptualize how you get comfortable with your positions without having some view on market conditions.
    Any insights would be most appreciated.

  26. Mark Wolfinger 06/03/2010 at 1:08 PM #

    I enjoy these back and forth questions. I don’t know why I cannot reply in a paragraph, but I cannot.
    I’ll post my reply asap – but it will probably be next week.

  27. Burt 06/06/2010 at 10:23 PM #

    Mark, I tried to post a reply on Thursday and now see that it did not appear. In any event, I enjoy this dialogue as well and look forwarrd to your reply at your earliest convenience.

  28. Mark Wolfinger 06/06/2010 at 10:39 PM #

    Reply in 7 hours (Monday morning, 6/7/2010)