How to calculate the expected move from a straddle (with a worked example)
You do not need a Black-Scholes calculator to get the expected move. The at-the-money straddle already contains it. This guide walks through the calculation by hand, shows where the popular shortcut goes wrong, and gives the correction that lines the straddle number up with the implied-volatility number.
Step 1: find the at-the-money straddle
Take the strike closest to the current stock price. Read the mid price (halfway between bid and ask) of the call and the put at that strike for the expiry you care about. Add them together. That is the straddle price: what it costs to own both directions.
Example: AAPL at 319.70 on a Friday close. For the expiry four days out, the 320 call is bid 2.05 / ask 2.15 and the 320 put is bid 1.85 / ask 1.95. Mid prices 2.10 and 1.90, straddle 4.00.
Step 2: express it as a percentage
This raw ratio is what many people quote as the expected move. It is close, but it is not one standard deviation, and that matters when you compare it with anything else.
Step 3: convert to one standard deviation
For a normally distributed price move, the theoretical value of an at-the-money straddle is
where σ√T is the one-standard-deviation move for the period. So dividing the straddle ratio by 0.7979, or multiplying by 1.2533, gives one standard deviation:
AAPL's expected move to that expiry is 314.68 to 324.72 before skew adjustment. Put skew tilts that slightly lower in practice; the calculator's published range for the same chain was 313.68 to 323.71.
Why the 0.85 shortcut understates it
The widely circulated rule is "expected move = straddle × 0.85". That gives 1.06% here. The rule is not made up: 0.85 (some use 0.8) is roughly the ratio between the mean absolute move and the straddle price, a quantity traders sometimes call the "average expected move". But it is about 0.68 standard deviations, not one.
| Quantity | Multiplier on straddle ratio | Share of closes inside |
|---|---|---|
| Straddle × 0.85 ("average move") | 0.85 | about 50% |
| Raw straddle ratio | 1.00 | about 57% |
| One standard deviation | 1.2533 | about 68% |
None of these is wrong on its own. The mistake is mixing them. The implied-volatility formula everyone uses for longer expiries (IV × √T) is one standard deviation. If a tool uses 0.85 × straddle for weeklies and IV × √T for monthlies, its short rows are quietly 32% narrower than its long rows for reasons that have nothing to do with the market. Pick one standard deviation everywhere and the rows become comparable.
Step 4: use more than one strike
When the stock sits between strikes, the single closest straddle is slightly biased. A better estimate weights every straddle within about 2% of the price by how liquid it is. OptionsMovement uses the smaller of the two legs' volume + 0.3 × open interest as the weight, so a straddle with one dead leg cannot dominate. On a liquid name like AAPL the difference is a few hundredths of a percent; on a thin name it can be material.
Step 5: price the legs carefully
Mid price is usually right, but not always. If the last trade is recent and the spread is wide, the last trade carries real information. If the last trade is stale (an hour or more old during market hours) it should barely count. A spread-aware mid that leans on the bid/ask when the spread is tight and on the last trade when it is wide gives a more honest straddle than either alone.
When to use the straddle method and when to use IV
Under about a week to expiry, use the straddle. Quoted implied volatilities for very short-dated options are computed from tiny time values and swing wildly with a one-cent change in price. The straddle is a real market price and does not have that problem. From about a week out, the two methods agree closely and the IV formula is simpler to reason about. The calculator switches at 8 days; both give one standard deviation, so nothing jumps at the switch.
A note on 0DTE
On the expiry day itself the straddle still works, with one twist: "days to expiry" is a fraction of a day and shrinks through the session, so the range shrinks with the square root of the time left. A range read at 9:30 is not the range at 2:00. The 0DTE guide covers that case.