I couldn't get filled on KMX and didn't want to hug the bid. Shorted 1 AZO but it weirdly dumped almost $10 in the last 10 mins before close. I have the $1155 straddle. So this could be really good or pretty bad depending on what the reaction is.
1155 seems ATM to me. It closed at 1154.xx I see the drop after hours, which is probably because of low liquidity. I was going to short AZO, but I was not keen on shorting an $80 straddle.Good luck with AZO Magic! Are you short OCT? Something I have been doing recently is buying the wings on the first-week expiration for a few pennies and selling the 30-day straddle. The 30-day straddle has the same area under the curve as the first week straddle but it has a larger width and lower peak which seems a better fit for earnings given we usually do see some type of move up or down.
Huh, here's my data. And yeah the price is high, was on the fence. Had a good day today though so the house money bias got me. Also we seem to be on the same page.. not with earnings yet but another strategy I just started running calc for best risk control to cost over various diagonals. Wings were really eating my expectancy earlier on earnings so in the face of a small edge I have dropped them and sized down drastically on most smaller names. Maybe there's a happy medium.
Interesting that you've stopped buying wings on these trades. I agree that buying wings erodes the edge, and also makes it more difficult to close out post earnings. But this strategy has a negative skew, so the edge that's eroded by buying wings would eventually be offset by the occasional explosive move. In fact, I've actually been leaning towards buying narrower wings lately. Besides, all our trading is manual, so the few cents of edge gained/lost on each trade is insignificant compared to our "luck" in timing the exits.
sup OM! You're right. For some reason on ToS the 1147.xx tick happened after market close. OM, has a good point, the trade has excess kurtosis so in the long run the wings might not be as expensive as we think. I thought of something that might be of interest. Selling Vol on companies with the highest implied move. Not relatively but in absolute terms. Since most fund managers mark-to-market. If a company has an implied move of 10% she is more likely to hedge her position vs a company that has an implied move of 2%. This could really drive up the premium for the options causing more fund managers to hedge their position... before I run any backtests, does that theory make much sense? On Bloomberg there is the earnings function ERN<GO> where the implied move number is up front and centre. I would imagine if AAPL had an implied move of 20% the day before their earnings, fund managers would almost be forced to buy vol.
Hey TBS...Glad to see the earnings journal is back up. I might end up posting trades this time, although I don't expect to make any earnings trades until mid October. If what you say is true, then companies with the highest implied moves would also tend to have the highest implied/actual moves (because fund managers bid up the vol even further). But I think the numbers say that the stocks with higher absolute implied moves tend to have proportionally higher actual moves, usually higher even than those with lower implied moves. That said, I was actually thinking about this too (short vol on stocks with the highest absolute implied moves). You might recall that I was looking to play these with double calendars but abandoned the idea when I found that the back month post-earnings vol has a lot of variance depending on the market reaction. Might still be worth looking into, although I haven't revisited it recently. I think a single long calendar at the ATM strike is probably the best way to play these. They have lower probability of success but generally offer much better return if the stock doesn't move at all.
Yeah, I’ve been trying to get my mind around different parts of the surface recently. On one hand wings are unsustainable from the vol figures. But on the other, they can’t dislocate too much from the rest of the surface. So in a long enough series they may very well be underpriced. But it could be years until the expectancy shows up. I tried to do some math to see how much negative skewness drags on geo returns. I mostly confused myself but I had some figures * sigma^3.. and the conclusions I was getting we’re showing that it didn’t have a severe effect on large enough series of periods. Still not sure what fair compensation for selling positive skewness should be. But if I had to pick a side I’d say the market overprices it. This kind of supports buying closer wings. Or dirt cheap short expirations for just buying a definite risk cap. The worst area to be is the one month 5 deltas or something. If margin wasn’t an issue I’d almost be thinking we should opportunistically be selling these; just over the crush and in appropriate size. — Regarding high implied, one of the things I was puzzled by is that earnings seem to generally normalize at a fixed spread slightly above or below the actual moves. I initially would’ve expected a wider margin on the more volatile ones because the shorts are taking on more variance. Like 9% implied over 8% avg move is a different trade than 3% implied over 2% avg move. Maybe because the risk can be diversified away easier it gets flattened? I think there might be something there in the high implied to exploit but if it was straightforward better vol edge wouldn’t we be seeing it in the avg spreads over actual move on these? I’ve wondered before if there are some mean reversion dynamics to exploit. Especially if we’re smaller and can hedge after hours, some of these big movers that do like 8% jump 6% intraday but avg out to 10% avg day before start becoming good candidates for longs. Edit - some details
I have always wondered. If you have to hit the bid to get filled, there is clearly no aggressive buyers. This could be a great thing.
This is a great conversation we have here. Why have we never thought about earnings as a long kurtosis trade? None of us manage over 50 million, we are eligible to trade about 1000 companies 4 times a year. That is way more than enough trades to realize any edge we have. 3 sd puts/calls before earnings are usually bid .05 ask .10. Take a look at ULTA 3sigma puts right before earnings. ATM Implied SD for ULTA was ~$25. So 3x 25 = 75. 337.5-75 = 262.5. The puts were going for pennies. The next day they were $30. I have seen countless 3 sigma moves during earnings and a fair share of 5 sigma moves this year (STMP, ULTA, ALGN, KRAFT, KO) just to name a few. If we can build a model that tells us what companies are more likely to have larger tails that would be great. I believe the math topic we are looking for is called "Extreme Value Theory". https://www.banqueducanada.ca/wp-content/uploads/2010/01/wp00-20.pdf