I wouldn't bother trying to decipher the numbers. The point I was trying to make is that the ATM call skew is highly variable with time. Here's a chart that will make the point clearly.
Betting on unflattening of the ATM IV curve
- Thread starter morganpbrown
- Start date
I wouldn't bother trying to decipher the numbers. The point I was trying to make is that the ATM call skew is highly variable with time. Here's a chart that will make the point clearly.
as the others said, trading ATMs is really hard because ATMs never stay ATM...you're betting against vol/spot correlation here.
FOMC tomorrow btw....just saying because "safe for a 1std move"
thumbs up for putting the trade up here, tho
So I'm stoked that the smart dudes responded. And I have not (yet) seen the word "dumb" used, so at least there's that. Maybe people are just being nice.
See below a screenshot of the P&L and net greeks vs DTE for 3 cases: price & IV unchanged, Price up/IV down, and Price down/IV up.
People kept telling me that negative gamma was bad, but it just wasn't sinking in. Now I get it.
Phi slamma gamma.When the price drops, I am piling on the risk (increasing delta from zero). When the price rises, I'm essentially shorting the stock (reducing delta from zero).
Gamma was so small, I thought to myself, it can't be THAT big a deal, right? But jeez, after 5 straight days of 2% negative gamma, and I'm sitting on +/- 10 delta.
The other greek that moves against me is theta. But when I look at the P&L, especially of the "price down/IV up" case, it's clear that theta alone isn't killing the P&L. I guess it's mainly gamma.
In any event, I can't see myself letting this trade run for more than another 7-10 days. It looks like the wheels don't truly fall off until 7 DTE.
Last edited:
first of all, at least you're doing your homework and unless you're behaving like a dick, there is no reason to be meanSo I'm stoked that the smart dudes responded. And I have not (yet) seen the word "dumb" used, so at least there's that. Maybe people are just being nice.
See below a screenshot of the P&L and net greeks vs DTE for 3 cases: price & IV unchanged, Price up/IV down, and Price down/IV up.
People kept telling me that negative gamma was bad, but it just wasn't sinking in. Now I get it.
When the price drops, I am piling on the risk (increasing delta from zero). When the price rises, I'm essentially shorting the stock (reducing delta from zero).
Gamma was so small, I thought to myself, it can't be THAT big a deal, right? But jeez, after 5 straight days of 2% negative gamma, and I'm sitting on +/- 10 delta.
The other greek that moves against me is theta. But when I look at the P&L, especially of the "price down/IV up" case, it's clear that theta alone isn't killing the P&L. I guess it's mainly gamma.
In any event, I can't see myself letting this trade run for more than another 7-10 days. It looks like the wheels don't truly fall off until 7 DTE.
View attachment 270466
Second, when you short front/long back your gamma/theta profile gets worse and worse because the front month decays to a point where there is almost no optionality left. I mean if all you get is 20 cts till expiration for a stock that moves 2$ a day, what's the point, right?
So: The higher vol and/or the more days to expiration, the higher theta and the lower your gamma. Hammer that into your brain. Much premium->not much gamma. Good for seller, bad for buyer, oki?
Short front/long backmonth is tricky because of that.
Shouldn't the theta also go down with gamma as we increase DTE?
I think @MrMuppet is talking about the total theta and gamma for the calendar, not a single option that's further out. Long calendars are +theta and -gamma, so where a 30D/60D ATM calendar in SPY is approx 0.025t and -0.007g, a 30D/150D is 0.045t and -0.015g. I.e., higher theta and lower gamma.
So the theta and gamma are both doubled comparing the 30D/150D against the 30D/60D calendar. There doesn't seem to be a gain. I thought @MrMuppet was trying to increase theta and reduce (the magnitude of) gamma.
Last edited:
Vega is not 'additive' across different expiries.
Should consider both Raw Vega and some form of Weighted Vega.
If you use SqrtTime to weight the Vega, the calendar spread is relatively Vega Neutral
Raw Vega = 0.86 - 0.40 = 0.46
Weighted Vega = 0.39 [0.86 x Sqrt(17/80)] - 0.46 = (0.07)
100% no. Think about it like this: An option has 100$ premium and 20 DTE. We now asume theta is linear (which it isn't but...you know, simplicity)
You collect 5$ per day.
If the option has 200$ premium you get 10$ per day, makes sense? Now what about gamma?
Think about a 100$ strike option that has 20 DTE and it's premium is only 2 cts. The stock is moving 1$ per day average. Meaning on a single day, the option could either have a delta of 1 or 0 at the end of each day. Your gamma is insanely high but you only collect 2cts over the next 20 days -> your theta is tiny.
Now if the 100$ strike option is worth 200$ premium (which is absurd for a 100$ stock, but...simplicity), at the end of the day the option delta is still 0.5 but you collect 10$ theta for that...per day. Gamma is basically zero
The important factor is the theta/gamma ratio: How much do I lose when the stock moves vs. how much do I get for it in decay.
The driving force is implied volatility. The higher implied vol, the better the theta/gamma ratio.
If you compare different maturities, you want to look at this ratio. Theta goes down with DTE since you have to wait longer to realize the premium but gamma is even lower.
When you trade calendars you want to capture curve rolldown. Meaning if front is lower than backmonth, you want to buy front month and sell backmonth. You buy cheap gamma, because IV low-> theta/gamma is better for the buyer and you sell expensive theta because IV high, theta/gamma is better for the seller.
Plus if 30 DTE is at 20 vols, 60 DTE 30 vols and 90 DTE 40 vols and you sell the 60/90 calendar, you buy 30 vols and sell 40 vols. As 30 days pass, your 60 DTE will now be 30 DTE so you lost 10 vols but you made 10 vols as the 90 DTE becomes 60 DTE. As the longer dated option has more vega, you win here, too.
BUT you're short vol and that's your risk.
When you look at what the majority of retail is doing with calendars: They wait until market does not move - which makes the front trade lower than the backmonth.
Then they sell cheap front month theta and buy expensive backmonth gamma.
In extreme situations you are short gamma and get zero theta for that, plus you're long vol and short rolldown. Not very smart
Last edited:
Man, that's a whole semester's worth of education on calendars - helped clarify several things I was fuzzy on. Thank you, MrMuppet!
However: guilty as charged on typical retail behavior (with, of course, the expected results.) So, what would you consider a reasonable setup/market state for opening a calendar trade? I'm trying to picture a situation in which you have relatively high front month theta and cheap back month gamma, and getting a confusing image of high IV in the front with little to no movement, and way out tenors in the back.