General Questions on the DITM Double Diagonal

[cwelsh 1,497]

For whatever reason, I did not have access to SO on Thursday-Saturday evening, I'll have updates and thoughts posted over the next couple of days (client meetings most of the morning here)

[tjlocke99 18]

Chris:

Seems like the DD discussions have gone a bit cold lately. I'm still in the VXX and GLD trades and following a couple of different approaches. To catch up, I think we were discussing using OTM longs instead of ITM. Have you reached any conclusions about best deltas? I have been fiddling with this and by my reasoning (possibly faulty) it actually looks like ATM works best. Here is my reasoning:

It seems to me that the purpose of this trade is to sell time value, choosing short deltas that balance credit received with the likelihood of getting hit. So, choosing deltas that will only get hit, what, maybe 1/2 to 2/3 of the time? Even if we get hit 3/4 of the time on one leg, but set stop losses so we retain SOME credit for the week (and we get whipsawed for losses very infrequently) then it works out. I have lost track of what deltas should be working best, since they change with each underlying. A recap would be helpful, if someone could provide it.

BUT, we don't want to risk of holding naked shorts, so we buy longs. The trick here is to find the longs that are the cheapest in the long run (i.e. least cost including both initial investment AND theta decay). With DITM, we don't have much theta, but the upfront cost is high, so lower ROI overall. With DOTM, basically same theta, and the lower initial cost is offset dollar for dollar by margin requirements. I think I would rather have the margin requirement, as I could put that money in something safe but still mildly profitable and increase my overall return a bit.

However, I have been fiddling with the numbers (so far, only on GLD) and it appears to me that ATM works best. If you plot the theta decay from week to week, out about 3 months, it doesn't vary much for OTM, ATM, ITM, DITM or DOTM as long as you sell them 3 weeks before expiration. With DOTM and DITM, you might be able to go another couple of weeks closer. (For GLD, the weekly theta decay averages about .17/week ATM, about .19 DITM or DOTM and .21 OTM or ITM) So if theta losses are similar, then it comes down to total initial cost, which is lowest ATM. I haven't tried this with real money yet, but it appears that the DD trade has the highest ROI with ATM longs.

Then it is just a matter of choosing the best shorts.

Chris, please shoot holes in this reasoning before I am tempted to think I have learned something meaningful. And let's continue the discussion, as I think this is a very viable trade option.

Kelly,

If you do ATM longs and ATM shorts you have a traditional calendar which Kim has traded here recently and Kim has a forum thread on it. If you do a double calendar with the puts and calls ATM they have the same P/L graph, so you are really just doubling your position.

If you go long a far dated position and sell a shorter dated position that is further OTM than the long than your position is a diagonal or double diagonal.

I believe the difference between these pretty vanilla option strategies and the idea we were discussing is the following. You short an OTM option but you also cover it with a further date OTM long. For example: with GLD closing at 164.44 today something like:

long 170 Feb call

long 159 Feb put

short 167 Dec4 call

short 162 Dec4 put

then you roll that weekly short every week.

I'm not sure how the backtesting turned out, but I think this was the jist of the idea. The shorts could stay at the .20 to .25 delta as were being done with the original trade and as for the longs they are probably a slightly higher delta - maybe the .30 to .35 delta range?

[Kelly Park 10]

Richard:

Actually, we started with DITM longs and OTM shorts. There was discussion of going to OTM longs also, but the conversation sorta died there. To me, the longs are just a way to "allow" us to sell shorts without being naked, so we are just looking for the cheapest longs possible, factoring in theta decay, purchase price and margin. But I'm not sure which deltas of OTM is best overall for the longs.

Another issue that is coming up now: There are opportunities to roll the longs out a month for no cost or even a credit. I would be interested in others thoughts about when to roll and how much credit makes it worthwhile.

[tjlocke99 18]

I haven't been as active on this forum the last few months, but I was wondering what happened to these trades which were changing to a variant of a double diagonal where the long, long month was actually further OTM than the short, short month.

 

Also, I haven't seen Chris posting recently.  Is he still around?

 

Thanks!

[fieldydwb 31]

Am interested in this as well.  Been away for a few months but am back now.

[urfiend 8]

Have you done / tested DITM double diagonals on indexes?

What is the advantage against the ATM double diagonals since they are cheaper and the real income is generated by the short near term options?

If you by a second ATM straddle you even profit from large moves of the underlying making up for the debit loss on the ITM diagonal and are able to adjust your center from time to time at a profit and cover yourself for a changing market direction!?

As I understand your rules for the SPY ratio diagonals are now in the anchor forum or can I find them anywhere else on SO?

Thank you for your advice and clarifying