Trading events

OpeningA straddle spells out the market’s expectation

XYZ, a $200 stock, reports earnings after the close tonight. The ATM 200 straddle expiring this Friday (2 days left) is quoted at $11.81, and IV for this expiration is 100%. You think this report will be a big one, so you buy 1 straddle.

The next day the stock jumps 5% and closes at $210. Neither the direction nor the size is small, yet the straddle is worth only $10.02, a loss of $178 per straddle. If the stock hadn’t moved at all, you would have lost $930.

The $11.81 you paid is the price the market set on this earnings move: 11.81 ÷ 200 = 5.9%. If the actual move is smaller than about 5.9%, the buyer loses and the seller wins; if it is larger, the reverse. Event trading really has only one question to answer: compared with how far this stock will actually move, is the move the market implies expensive or cheap? This chapter first teaches you to read the implied move, then to strip it down cleanly, and only then gets to structures, macro events and 0DTE.

10.1The earnings IV cycle

IV climbs all the way into earnings and collapses the moment the report is out. Most of this cycle comes from arithmetic; market sentiment is just a layer on top.

Split a stock’s variance into two pieces: the base variance it has every day, plus one extra jump variance on the night of earnings. Any option that expires after earnings has to pay for that jump.

The key is the v_e / T term. The size of the jump stays the same, but the closer you get to expiration, the smaller T becomes; the same jump is spread over fewer days, so it translates into a higher annualized IV. That is why front-month IV climbs on its own even if nobody changes their view of the earnings. Once the report is out, the jump has happened, v_e drops out of the formula, and IV falls straight back to σ_b.

Some numbers: base volatility is 30%, the market expects a 7% earnings jump (1 standard deviation), and the report comes out after the close on day 21. The front-month option expires on day 25, the back month on day 53.

Three ways to read it:

• The front month rises a lot, the back month a little. An inversion with front-month IV above the back month (the term structure from Chapter 6) is in itself evidence that the market is pricing an event.
• After the report, the front month falls hardest. In the table, the front month drops from 73.3% to 30%, a fall of 43 points; the back month drops only 8 points. A calendar spread that sells the front month and buys the back month earns exactly this difference.
• The real world adds another layer of demand. Speculative buying pushes IV higher than the model, and during earnings season IV often starts climbing weeks ahead. The model tells you “where IV should be if the expected jump stays the same”; only the part above that is sentiment.

10.2Implied moves and historical moves

The price of an ATM straddle is roughly the “average absolute move” the market expects, and a 1-standard-deviation move is about 1.25 times that.

The derivation takes one line. ATM call and put prices are each about 0.4 × Sσ√T (the rule of thumb from Chapter 3), so the two together are about 0.8 × Sσ√T. And the mean absolute value of a normal distribution is exactly √(2/π) ≈ 0.798 standard deviations. The two 0.8s are the same number, so:

You can also read off a probability that often gets overlooked: under the normal assumption, the chance that the actual move exceeds the straddle price is 2 × (1 − N(0.798)) = 42.5%. A straddle buyer makes money at expiry less than half the time, but wins big when winning; the seller is the exact opposite.

What to compare it with

For the historical move, use the percentage change from “the last close before earnings” to “the first close after earnings,” because options are also valued at the close. Take the last 8 reports and look at two numbers: the mean and the median of the absolute moves. Below is hypothetical data for XYZ over the past 8 quarters.

Mean absolute move 4.28%, median 3.6%, and only 2 of the 8 moves exceeded 5.9%. Implied ÷ historical = 5.90 ÷ 4.28 = 1.38: the market’s price is nearly 40% more expensive than the average of the past.

Where the ratio thresholds come from

A common rule of thumb: consider selling only when implied ÷ historical is above 1.2, consider buying only when it is below 0.9, and do nothing in between. The thresholds come from sampling error. The standard deviation of the 8 absolute moves is 2.40 percentage points, so the standard error of the mean is 2.40 ÷ √8 = 0.85. In other words, the “historical mean of 4.28%” could itself be anywhere from 3.4% to 5.1%, which puts the ratio anywhere from 1.15 to 1.72. When the ratio is only 1.1, the “expensive” you see may well be just noise.

The thresholds don’t apply when the company has just been through a major change (a merger, a new CEO, a turn in the industry cycle), so the last 8 reports no longer represent the future; when one extreme value in the history dominates the mean; or when the options are illiquid and the straddle’s bid–ask spread alone is more than a tenth of its price.

10.3Isolating event IV

The IVs of two expirations are enough to solve for two things: the base volatility of ordinary days, and the size of the jump on earnings night.

A straddle’s price mixes in the time value of ordinary days. With 2 days left in the front month, that is not a big problem; with 10 days left, a lot more gets mixed in. The clean approach is to take two expirations that both include earnings and solve the equations from Section 10.1. Total variance w = σ²T is a straight line in T: the slope is the base variance and the intercept is the jump variance. Two points determine a line.

A worked example

XYZ at Wednesday’s close: the Friday expiration (T₁ = 2 days) has IV of 100%, and the next Friday’s expiration (T₂ = 9 days) has IV of 54%.

1. Total variance: 1.00² × 2/365 = 0.005479; 0.54² × 9/365 = 0.007190.
2. Base variance: (0.007190 − 0.005479) ÷ (7/365) = 0.0892, σ_b = 29.9%.
3. Jump variance: (1.00 − 0.0892) × 2/365 = 0.004991, event 1σ = 7.06% ($14.13), expected absolute move 5.64% ($11.27).
4. Event-day IV: √(0.0892 + 365 × 0.004991) = 138%, 4.6 times that of an ordinary day.

Read straight off the straddle, the move is 5.90%; with the event isolated, it is 5.64%. The 0.26-point difference is the time value of the two ordinary days. When comparing with history, use 5.64%: the ratio is 5.64 ÷ 4.28 = 1.32, still on the expensive side, but not as extreme as 1.38. After the report, front-month IV should reasonably land at σ_b ≈ 30%, which is the “back down to 30%” assumed in the opening table.

10.4Choosing an earnings structure

The order for choosing a structure: look at the implied ÷ historical ratio first, then set the maximum loss you can take, and only then pick the specific legs.

Below, five common structures are placed in the same scenario: you open at XYZ’s Wednesday close, with front-month (2 days) IV at 100% and back-month (9 days) IV at 54%; you value the position at Thursday’s close, with IV for both expirations back down to 30%. The iron condor and reverse iron condor use strikes 190/180 and 210/220; the calendar sells the Friday 200 call and buys the next Friday’s 200 call.

±5.6% is the expected absolute move isolated in Section 10.3, and ±8.9% is XYZ’s largest move in its history. Note three things when you read this table:

• The short straddle makes the most when the stock “doesn’t move” ($930), but at ±8.9% it loses $600 in one go, and at −15% it would lose $1,817.
• The short iron condor makes only $319 when the stock doesn’t move, but its breakevens are actually wider than the straddle’s (the short legs are beyond ±5%, plus the premium collected), and its maximum loss is locked at $681.
• The calendar’s P&L is small either way, and its maximum loss is the $93 paid. It is very sensitive to IV after the report: if post-report IV is 25%, it makes only $105 when the stock doesn’t move; at 40%, it makes $220.

Entry timing, strikes and exit

• Entry timing. Sellers usually open the position just before the close on the last trading day before earnings: that is when front-month IV is highest and the fewest ordinary days remain. Opening a few days early means carrying a few more days of ordinary-day Gamma, and you may watch IV keep climbing.
• Where to put the short legs. The short legs of the iron condor in Table 10-3 are at ±5%, closer than the expected absolute move (5.64%); that is an aggressive setup. Under the market’s implied distribution, the chance of a move larger than 5% is about 50%. Move the short legs out to 185/215 (about 1σ away) and that chance falls to about 31%; the price is that the credit drops from $319 to $184 and the maximum loss rises from $681 to $816. There is no free margin of safety, only different trade-offs.
• Exit. By default, close the whole position 15–30 minutes after the open on the first day after the report. The bid–ask spread is widest at the open, so wait for it to narrow; by then IV has already been crushed, and the money the event trade was meant to make has been made. Letting an event position drift into an ordinary holding means swapping in a trade you never analyzed.
• Handle a calendar as one unit. As the front-month leg nears expiry, either close the whole position or deal with the front-month leg early, so you don’t get assigned and end up with a lone back-month leg.

10.5Macro event days

A macro event hits an index far less than an earnings report hits a single stock, but it hits every position in your portfolio at the same time.

Why VIX barely moves before CPI

Work it through with the framework from Section 10.3. Suppose the base volatility of the S&P 500 is 14%, and the market expects an extra 1% jump (1σ) on CPI day.

• Options expiring in 1 day: IV = √(0.14² + 0.01² × 365 / 1) = 23.7%, almost double.
• 30-day IV (the VIX measure): √(0.14² + 0.01² × 365 / 30) = 14.43%, only 0.4 points higher.

The same jump looks huge over 1 day and gets diluted away over 30 days. So reading VIX to judge “whether the market fears CPI” badly understates the event premium; look at straddles expiring the same day or the next day instead. With SPX at 7,800, a 1-day ATM straddle costs about $45.6 (0.58%) on an ordinary day and about $77.2 (0.99%) on this CPI day.

Hedging before an event: expensive per day, cheap per event

Suppose your portfolio is equivalent to 300 shares of SPY (beta-weighted, as in Chapter 4), and you want to add protection the night before CPI, with SPY at 780.

• 3 of the 780 puts expiring the next day, at 23.7% IV: $382 each, $1,145 in total, covering just this one day.
• 3 of the 30-day 780 puts, at 14.43% IV: $1,161 each, $3,483 in total, or $116 a day.

Per day, the one-day hedge is nearly 10 times as expensive (1,145 ÷ 116 ≈ 9.9), because its price is almost entirely event premium; the upside is that you pay only for the one day you need. Conversely, if you were planning to hold a month of protection anyway, buy the 30-day puts before the event: the event premium is only a small part of their price.

Index IV after the event

When the result is no surprise, short-term index IV usually falls back after the release, for the same reason as with earnings: the jump has happened. The difference is that the drop is smaller, and a macro event can change base volatility for the following weeks. For example, if the FOMC signals a new policy direction, the market may keep repricing over the next few days, and base volatility can actually rise.

The real risk of these events is at the portfolio level: every stock moves in the same direction at once, and diversification across stocks basically stops working for that day. Before the event, check your portfolio’s beta-weighted Delta (Chapter 4) and, if needed, add an offsetting position with short-dated index options. During the event, do as little as possible; wait until the press conference is over and prices have settled before acting.

10.60DTE: a whole day compressed into a few hours

0DTE options expire the same day. They squeeze a whole day of risk into a few hours, pushing both Gamma and Theta to extremes. SPX, SPY, QQQ and other major index products now have expirations every trading day.

A few numbers worth remembering:

• ATM Gamma in the last hour is 11.7 times that of a one-month option, and 23.5 times in the last 15 minutes.
• An ATM option’s value is proportional to the square root of the time remaining. The $3.00 at the open is down to $1.16 by the last hour, about √(1/6.5) = 39%: time decay is slow at first and fast at the end, and the last hour eats nearly 40% of the opening value.
• In the last hour, a 0.2% rise in SPY ($1.56) takes the ATM call from $1.16 to $2.11, a gain of 82%; the same move at the open lifts it only 29%.
• For the 784 call, 0.5% out of the money, Gamma peaks with a little under 2 hours to the close, then decays quickly, down to just 0.006 in the last 15 minutes: the option either goes to zero or gets “caught” by the stock and becomes at the money.

How institutions and retail traders use them

Its effect on intraday price action

• Intraday Gamma is concentrated in the last two or three hours, and so is market makers’ hedging flow. When they are long Gamma, prices are more likely to get “pinned” near big strikes; when they are short Gamma, moves are more likely to accelerate (Chapter 4, Section 4.8).
• Most 0DTE positions are opened and closed the same day. Open interest, updated only once a day, doesn’t see these positions, so Gamma estimates based on OI miss them.
• Published research has not reached a consensus on whether 0DTE systematically amplifies market volatility. The safer view is that it has changed the distribution of intraday volatility, especially on event days.

10.7SPX and SPY: one index, two contracts

SPX and SPY options track the same S&P 500, but they are completely different in exercise style, settlement and tax treatment, and in event trading those differences turn directly into money.

XSP is a mini SPX, one-tenth the size of SPX, with a notional value close to SPY’s. It is also European-style and cash-settled, and Section 1256 applies to it as well.

How much Section 1256 saves

Suppose your net options gain for the year is $10,000, the short-term capital gains rate is 32% and the long-term rate is 15% (assumed rates, only to show the calculation).

• SPY: all of it short-term, 10,000 × 32% = $3,200.
• SPX: 10,000 × (60% × 15% + 40% × 32%) = 10,000 × 21.8% = $2,180, or $1,020 less in tax.

Your actual tax depends on your personal situation and where you live; rely on the advice of a tax professional.

Two kinds of expiration risk

• SPX AM settlement. Standard SPX contracts expiring on the third Friday of the month stop trading at Thursday’s close, and the settlement price is calculated from the opening prices of the index components on Friday. If you hold them through Thursday night, Friday’s opening gap directly sets the settlement price, and there is no way to close in between. The weekly and daily SPXW contracts are PM-settled at the closing price and don’t have this problem.
• SPY after-hours exercise. Expiring SPY options stop trading at 16:00, but holders still have a window after the close to decide whether to exercise. Say SPY closes on Friday just below the 780 call you sold, and an after-hours headline pushes it to 783: the other side may exercise, and on Monday morning you open with a short stock position.

How to choose: for large notional size, when taxes matter, or for event spreads, prefer SPX; for a small account, finer position sizing, or when you actually want to deliver or receive shares, use SPY or XSP. When you sell ITM SPY calls, watch out for early assignment before an ex-dividend date.

10.8VIX options and recurring events

VIX options are priced off VIX futures, and they settle in an unusual way. Monthly expiration, triple witching and index rebalancing are liquidity events written on the calendar well in advance.

Three things that make VIX options different

1. The underlying is a future. The VIX index itself can’t be traded; VIX options are priced off the VIX future for the same expiration month. In calm periods the future usually sits above spot: with spot VIX at 15, the same-month future may be around 17, and whether a call is in or out of the money depends on the future.
2. The settlement price is a special opening quote. VIX options are European-style and cash-settled. On the morning of expiration (usually a Wednesday), a special quote (ticker VRO) is calculated from the opening prices of SPX options and used as the settlement price. It can be very different from the previous day’s closing VIX, and at the close the day before expiration you don’t know what the settlement price will be.
3. Expiration follows SPX. VIX settlement is generally 30 days before the next month’s standard SPX expiration, so it is not the third Friday of the month. Mark the VIX settlement Wednesdays separately on your calendar.

VIX is strongly mean-reverting: after a spike it often falls back quickly. That makes VIX calls a form of protection you “have to cash in on time,” which Chapter 11 covers in detail.

Events already on the calendar

Price behavior around monthly expiration is tied to Charm and Vanna from Chapter 4: near expiry, the Delta of OTM options quickly goes to zero, and the stock held as a hedge gets unwound; after expiry, the positions that had been “pinning” the price are gone, and the next week’s moves are often freer. The practical rule is simple: on these days, don’t hold a large short-Gamma position that is near expiry and near the money going into the close; in the first week after expiry, take a fresh look at the term structure and the Gamma distribution.

10.9Event calendar and IV dashboard

Write each event as one row of a table, and fill in the implied move, historical move, structure, maximum loss and exit rule before you place the order. A row you can’t fill in is a trade you shouldn’t make.

This is one of the deliverables of this course: an event dashboard you update every Sunday. The template below fills in three rows with this chapter’s examples. The numbers for XYZ and ABC come from the isolation formula in Section 10.3, and the SPX row comes from Section 10.5.

Each column matches a checkpoint from this chapter:

• Front / back IV: used to isolate the event, and also to check whether the term structure is inverted.
• Straddle-implied and isolated event move: the first is a quick estimate; the second is what you compare with history.
• Ratio: above 1.2, consider selling; below 0.9, consider buying; in between, leave it blank. ABC’s 0.79 says the market’s price this time is cheaper than the stock’s average past move.
• Max loss: fill in this column first, then work out the number of contracts. Keep the maximum loss on any single event to no more than 1%–2% of the account.
• Exit rule: an event trade is a bet on one night. Closing on the first day after the report is the default; holding on needs a new reason.

You can add more columns: IV Rank (Chapter 6), the straddle’s bid–ask spread as a share of its price (walk away above 10%), whether the earnings date is confirmed, and whether any other event falls in the same expiration.

CasesThree cases

ToolsChecklist and self-test

Your checkmarks are saved only in your own browser.

Five self-test questions