riskarb: Thanks for the response. Not sure I understand what you're saying. The loss on adjustment is the debit incurred by rolling down from 1140/1150 to 1130/1140 at a particular strike price of the SPX index. Whether you'd actually do that or not is a function of a number of factors. At a higher volatility the price of each option is higher, but with the spreads we're looking at differences. And here I'm looking at differences between differences (in the form of the spread). These curves were generated by calculating option prices for the 1150, 1140 and 1130 strike prices using the B-S model. At higher volatility, the vertical distance between the curves is less than at lower volatility. Therefore, theoretically at least, it seems that the debit incurred by rolling down is less than at lower volatility. Again, whether you'd actually want to do this or not is dependent upon a lot of factors. The curves don't emphasize this but, the point of adjusting is to establish a rolled down position that hopefully will stay OTM until expiration. These curves are included only to generate some discussion with regards to adjusting not to suggest that adjusting can occur over the entire range shown. Therefore, the severe ITM portion of the curves isn't practical. This is also true for the high OTM portion of the curve. Why would you adjust if you're 40 or more points OTM. Simply included to provide some perspective at each end, but not practical. When I said 'not nearly as great as I would have expected' I was talking about the debit to adjust. As I stated earlier, I'm not yet considering how much credit was brought in or rolling the calls down. I always thought that if I adjusted a position when the SPX price was 15 to 20 points away from the short, it would cost much less (i.e. 2 to 3 times less) than if I adjusted when the distance is 0 (ATM) to 5 points away. Since both the 15 point distance and the 5 point distance (or ATM) are both in the linear portions of the curve, this assumption that I had doesn't appear to be true. Use of this information is for those who actually trade credit spreads. I know you don't and you have made some good arguments for not. You're right that typically the adjustment debit shown by the curves would be 200 to 300% of the initial credit taken in. Countering this would be any credit taken in for a corresponding bear call spread which might also be rolled down to increase the credit further. In these postings, I'm simply trying to get a handle on just how much risk we are taking on.
I am not trying to dissuade any of you[yes I am!], rather, I am trying to point-out the advantage in taking an initial-hedge, rather than taking massive losses and rolling into an increasing stat-distro and adding risk by selling another round of cheap gamma.
I'm still learning the greeks and haven't developed the ability to think in terms of the greeks alone in making decisions. Could you provide a practical example of the intial hedge you're suggesting? Also, if you were to trade a credit spread, when would you add the hedge position?
I'm not trying to get hung-up on the greeks, but they're symbolic of the risk-curve. Gamma is not expressed in dollars, so it's tough to grasp for everyone. The time to hedge should be when you can sell-deltas and buy volatility cheaply, namely, in a deep otm long time spread. For example: Sell the 1120//1130 put spread at $.80 * 100 ... $8,000 credit, $92,000 req/risk. Buy 20 Nov/Dec 1135 put spreads at $5.20 ... $10,400 debit. As you can see you've paid a debit for the entire position, but the convergence gains favor the credit spread unless we trade substantially higher. You can certainly trade the time spread as 10 quantity and still retain a decent credit. This is only mitigating a small portion of the risk -- therein lies the problems with otm credit spreads; it's virtually impossible to get out cleanly when you need to. Another recommendation would entail replicating an otm, weak synthetic straddle with short futures. Sell the put spread and sell futures. Solve for the futures-gain at your short strike = your initial credit. Again, it's only going to earn you $8,000 on the futures hedge, but it makes the position PnL more symmetrical.
Here are my results for the October period, my first using this strategy. Call Side: 9/6 Sold 10 Oct 1290/1280 call spreads for 1.15 (1150.00) 9/22 BTC Oct 1280 Call for .40 (-400.00) 9/29 Sold Oct 1250/1260 Call Spread for .80 (800.00) The 1250 spread and 1290 calls expired worthless. Net Credit = 1550.00 Put Side: 9/13 Sold 10 Oct 1190/1180 put spreads for 1.00 (1000.00) 10/6 Closed 1190/1180 Put Spread for 3.00 (-3000.00) 10/7 Sold Oct 1140/1130 Put spread for .40 (400.00) The 1140 spread expired worthless Net Debit = 1600.00 Result for Oct is a loss of $50.00 When I put on the 1190/1180 P.S. the short strike was 50.00 OTM. When I closed it on 10/6 the market had fallen so far so fast that I was not comfortable rolling down because I did not see a bottom to the market. If I had rolled down 10 or 20 points, I would have been forced to adjust a second time for more losses. Getting out cooled my emotions and the next day I felt better about opening the 1140/1130 spread. I think I handled adjusting this about the best I could at the time. A mistake I recognize was putting on the 1190/1180 spread when the market was peaking. Had I waited a few days for a pullback, I would have put on safer strikes that I may not have had to adjust. So even though I had a loss for my first month with this strategy, I feel it was a good learning experience that will help me be successful with it in the future. Thanks to Phil and everyone contributing to this discussion for furthering my options education.
It is fine if this was your first time. If the strategy is completely new then certainly try to be as conservative as possible. No one likes losses but $50 is certainly easy to swallow for the experience. Just try not to trade too far in front of you or risk more capital than you are willing to trade with. Phil
Since the real risk in OTM credit spreads is the delta/gamma, is a volatility spread really the best approach to hedging? Increases in vol are not as severe a risk factor as gamma is so in my opinion any hedge that can add some deltas is preferably. As you pointed out, there is no perfect hedge and that is why strike selection and time to expiration are so important to allow theta to counter delta as best as possible in slow moving markets and stay out in front of the market move as best as possible in fast moving markets. The adjustments are basically to move out in front of the train and let time be an ally but naturally a collapsing market requires cutting it off at some point and taking the loss which can be mitigated through partial hedges and playing the other side. That is why I think long spreads or outright long options are the to of the list for hedging although expensive at times.
Coach, Going back to the SPY hedges. Suppose I had a 1160/1150 Put Spread and received a credit of $1.20. With SPX at 1,179.59, let's say I felt the market was going a little lower. And let's say I buy a 116/115 Bear Put Spread, for $0.30. But you said that if there is not that much time left, you would buy a butterfly. To convert that the above mentioned Bear Put Spread into a Butterfly, would I Buy another 116 Put and sell and another 117 Put? Sorry for bombarding you with so many questions. Thanks again.