Probability of Touch vs Probability of Expiring In the Money: What SPX 0DTE Traders Get Wrong
Probability of touch is about 2x the odds of finishing in the money. The math, SPX 0DTE tables, the skew catch, and stops and targets.
Key takeaways
- Probability of touch asks whether price trades at a level at any point before the close. Probability of expiring in the money asks only where price finishes.
- For a level away from the market, the touch probability is roughly twice the probability of finishing beyond it. Without skew, a 0.20 delta strike has about a 20% chance of finishing in the money and roughly a 40% chance of being touched before the close.
- SPX skew changes the numbers on the put side. On same-day SPX options a 0.20 delta put is priced at closer to 15 or 16% to finish in the money, so its touch probability is nearer 31% than 40%. Delta is a shortcut; the option prices are the source.
- Any rule that fires on a print, such as a stop order, an alert or an early exit, is a touch question. Pricing it with a settlement probability counts only about half of the times it triggers.
- The doubling rule has limits: drift, gaps and jumps, skew, which session counts as a touch, and the gap between option-implied odds and realized frequencies.
- Probability alone does not create an edge. A trade needs a better probability than the one priced into it, or a payoff that compensates for the odds.
The question behind the question
A trader sells an SPX 0DTE put spread with a short strike at 0.20 delta. The broker's platform shows an 80% chance of profit. By 11:15 the index has tagged the short strike, the trader exits for a loss, and the index recovers to close well above it. The platform was not wrong. It answered a different question from the one the trader's stop was asking.
Two questions get mixed up constantly in intraday options and futures trading:
- Where will price finish? This is the probability of expiring in the money (ITM), also called the terminal or settlement probability.
- Will price visit this level at any point on the way? This is the probability of touch (POT), also called the first-passage or barrier probability.
SPX options are European-style and cash-settled, so a position held to expiration pays on the settlement value alone. Everything else a day trader does with a level, including stops, profit targets, alerts and early exits, depends on the path.
Definitions
All three are usually computed from option prices, which makes them risk-neutral probabilities: prices of payoffs, not frequency counts of what has happened. We return to that limit below.
| Term | Question it answers | Typical source |
|---|---|---|
| Probability of expiring ITM | Will price finish beyond the strike at the close? | N(d2) in Black-Scholes, close to delta for a same-day option; with index skew, read it from how option prices change from one strike to the next |
| Probability of touch | Will price trade at the level at any time before the close? | About 2 x the probability of finishing beyond the level, for a driftless underlying with a continuous path |
| Probability of profit | Will the position finish beyond its breakeven? | The probability of finishing beyond the breakeven price |
Why touch is roughly double: the reflection principle
Take a level above the current price and a random walk with no drift and no jumps. Every path falls into one of three groups:
- Paths that never reach the level.
- Paths that reach it and finish beyond it.
- Paths that reach it and finish back on the near side.
The reflection principle says groups 2 and 3 have the same probability. For every path that touches the level and falls back, there is a mirror image, reflected at the moment of first touch, that touches and continues beyond. So the probability of touching equals the probability of finishing beyond, counted twice.
In symbols, for a driftless underlying and a level above the current price:
P(touch) = 2 x P(finish beyond the level)
For a level below the market, the same relationship applies on the downside, with the probability of finishing below it. This is a standard result, covered in the Quantitative Finance Stack Exchange discussion of barrier probabilities and in educator references such as options.wiki and the SPXXL glossary entry on probability of touch.
How to calculate probability of touch for an SPX day trade
- Take the distance from the current price to your level, in points. Strictly, measure from the forward price the options imply; on a same-day option it sits within a few points of spot.
- Estimate the standard deviation of the move over the time remaining, in points. A common starting point is the implied daily move scaled by the square root of the fraction of the session left. That assumes the day's movement is spread evenly; in practice the open and the close carry more of it.
- Divide distance by that standard deviation to get z.
- The finish-beyond probability is 1 minus the standard normal CDF of z. Double it for the touch probability, capped at 100% for levels very close to the price.
Worked example (illustrative levels)
Assume SPX is at 6,500 with a one-day standard deviation of 0.9%, which is 58.5 points. These levels are fictional and chosen to show the arithmetic, and the example has no skew.
| Distance from price | Points | P(finish beyond) | P(touch), about 2x |
|---|---|---|---|
| 0.50 standard deviations | 29 | 30.9% | 61.7% |
| 0.75 | 44 | 22.7% | 45.3% |
| 1.00 | 58.5 | 15.9% | 31.7% |
| 1.25 | 73 | 10.6% | 21.1% |
| 1.50 | 88 | 6.7% | 13.4% |
| 2.00 | 117 | 2.3% | 4.6% |
Read the 1.00 row carefully. The edge of the one-standard-deviation range is touched about one session in three under this model, even though price finishes beyond it only about one session in six on each side. The range describes where price ends up. It does not describe a corridor price stays inside.
Connecting it to delta
For a same-day option, delta is close to the probability of finishing in the money when the smile is flat. At about 0.84 standard deviations the finish-beyond probability is 20%, which is a 0.20 delta strike, and the touch probability is roughly 40%.
SPX does not have a flat smile. Downside strikes carry higher implied volatility than upside strikes, and that slope changes the probability the option prices imply. The probability of finishing below a strike is the rate at which put prices rise from one strike to the next, and with SPX skew it comes out below the put's delta. On SPX same-day options in October 2026, puts at 0.10, 0.20 and 0.30 delta were priced at about 7%, 15 to 16% and 25% to finish in the money, while calls were at or a little above their delta. For a 0.20 delta short put, that means roughly an 84% chance of finishing above the strike and a touch probability nearer 31% than 40%. The platform's "80%" in the opening example is the delta shortcut, not the market's own odds.
How touch probability changes through the session
Time decay shrinks the distribution of remaining moves, so touch probabilities for a fixed level fall quickly as the session runs. Using the same fictional setup, for a level 50 points above the market:
| Share of session remaining | Remaining standard deviation (points) | P(finish beyond) | P(touch), about 2x |
|---|---|---|---|
| 100% (the open) | 58.5 | 19.6% | 39.3% |
| 75% | 50.7 | 16.2% | 32.4% |
| 50% | 41.4 | 11.3% | 22.7% |
| 25% | 29.2 | 4.4% | 8.7% |
| 10% | 18.5 | 0.3% | 0.7% |
This table assumes constant volatility through the day. Real intraday volatility is higher near the open and the close, and it jumps on scheduled data and headlines, so treat the early rows and the last rows with extra caution.
Two practical readings follow. A strike that looks safely far away at 9:30 can carry a touch probability near 40%, which is why early-session stops on short strikes trigger more often than the settlement odds suggest. And the same strike has a small touch probability in the final hour, which is why many range-bound structures are placed later in the session.
Where the 2x rule breaks
The approximation is a useful sanity check. It is not a precise number. Five conditions matter.
- Drift and the forward. The argument needs a driftless walk. Under option pricing the reference point is the forward, not spot, and the two differ by carry. Measuring distance from spot and then applying a no-drift rule mixes the two. The effect is small over one day and larger over weeks.
- Gaps and jumps. Continuous-path math assumes price cannot skip a level. Prices gap through levels around news and at the open, so real touch frequencies can differ from the model. A stop order placed at a level triggers on a gap through it, even though price never traded at the level itself.
- Skew. Index options price downside and upside differently, so the finish-beyond probability on each side differs. Apply the doubling to each side with its own probability, read from the option prices, not to a symmetric band or to delta. The tastylive research team found that SPX moves fell inside option-implied expected ranges more often than theory predicts and traced part of the difference to volatility skew; see their study of two ways to measure expected move.
- Which session counts as a touch. SPX cash trades during regular hours. ES futures trade almost around the clock. A level hit in overnight ES may never print in the cash index. Decide which series and which session you are watching before any touch probability means anything. A single print in a thin market counts as a touch under the math, and may not count for you.
- Where the clock starts. A touch probability computed before the open, from overnight option or futures prices, starts the path at the overnight price. The cash session starts at the open. A level between the two can be crossed overnight and never trade in the cash index, so levels near the pre-market price are touched less often in the cash session than a pre-market calculation says.
Risk-neutral odds are not realized frequencies
Every probability above is derived from option prices. Option prices include a risk premium, heaviest on downside protection, so the priced odds of a large decline run above how often such declines have occurred. Doubling a risk-neutral settlement probability gives a risk-neutral touch probability with the same premium inside it.
For 0DTE the problem is sharper. Option Alpha's documentation for its 0DTE tooling notes that probabilities and expected value are less accurate for 0DTE trades because the time component in the pricing model is so short, and it recommends checking backtested, realized outcomes alongside the theoretical numbers (Option Alpha 0DTE Oracle documentation). The practical rule is to treat model odds as a reference frame and compare them with how often similar levels were actually touched in your own sample.
Turning touch odds into a trade plan
Touch probabilities matter most when you set a stop and a target. For a driftless market with no deadline, the chance of reaching a target before a stop is set by their relative distances:
P(target first) = stop distance / (stop distance + target distance)
| Stop (points) | Target (points) | P(target first), no deadline | Breakeven hit rate needed |
|---|---|---|---|
| 10 | 10 | 50.0% | 50.0% |
| 10 | 20 | 33.3% | 33.3% |
| 10 | 30 | 25.0% | 25.0% |
The two columns match in every row. In a driftless market, widening the target lowers the odds of reaching it by exactly the amount that raises the payoff, so expected value stays at zero before costs. This is the plain-math version of a point made on options.wiki: the arithmetic of expected value cannot produce an edge that the probability input does not already contain.
A same-day trade has a deadline: the close. Some sessions reach neither the stop nor the target before it, so each of the two chances is lower than the table shows, while the ratio between them stays the same. A plan that holds to the close also needs the odds of where price finishes between the two levels.
An edge needs one of two things: a better estimate of the odds than the one the market prices in, or a market that is not driftless over your holding period. Without one of them, spread, commissions and slippage make the expectation negative. That is the reason to be skeptical of any tool or strategy that quotes a high win rate without the payoff ratio next to it.
Which probability does each of your decisions need?
| Decision | Operative question | Use |
|---|---|---|
| Hold a cash-settled SPX spread to expiration | Where does price finish? | Probability of expiring ITM |
| Place a stop on a short strike | Does price visit the level? | Probability of touch |
| Set a profit target | Does price visit the target before the stop, and before the close? | Touch probabilities for both levels, with the close as a deadline |
| Judge a range-bound structure | Does price stay between two levels all session? | Touch probabilities for both edges |
| Alert or order triggered on a print | Does price print at the level? | Probability of touch |
How morphic.trade approaches it
morphic.trade publishes a pre-market probabilistic session map for SPX index options and ES futures. It reads that day's option prices and market positions and gives the probability that each level is touched, along with where the close may land, in plain English. The odds come from the day's option prices, skew included, not from delta. The map is published before the bell, is not redrawn afterward, and any re-issue is marked on the page. Every map is graded against the session it described. It is a map, not a trade signal.
Subscribers can test an entry, stop and target against that day's map: the plan runs on the day's priced distribution, so it reports the chance the target comes first, the stop comes first or neither by the close, the hit rate the plan needs, and the expected value. morphic provides market-structure commentary, not directional trade calls or investment advice.
Frequently asked questions
What is probability of touch in options?
It is the probability that the underlying trades at a given price at any point before a stated time, usually expiration, whether or not it is still there at the end. It is a question about the path of price.
Is probability of touch double delta?
Roughly, for an out-of-the-money level, a short horizon and a flat smile. Delta is close to the probability of finishing in the money, and the reflection principle puts the touch probability near twice that. On SPX the skew breaks the shortcut on the put side: a 0.20 delta put is priced at about 15 to 16% to finish in the money, so its touch probability is nearer 31% than 40%. Use the probability the option prices imply, then double it.
What is the difference between probability of touch and probability of expiring in the money?
Probability of expiring in the money looks only at the closing price. Probability of touch looks at every price traded along the way. The touch probability is always at least as large, because any path that finishes beyond a level must have crossed it.
How do I calculate probability of touch?
Divide the distance to the level by the standard deviation of the move over the remaining time, compute the probability of finishing beyond the level, and double it. For SPX, take the finish-beyond probability from the option prices so the skew is in it. The worked tables above show the arithmetic.
How often does SPX reach the edge of its expected move during the day?
Under the model, the edge of a one-standard-deviation range is touched about one session in three. Realized frequencies differ because of skew, gaps and risk premium, so compare the model with your own history before relying on it.
Does probability of touch apply to stop-loss placement?
Yes. A stop is a path question, so the touch probability of the stop level estimates how often the stop triggers. A stop inside the market's normal intraday range will trigger frequently regardless of whether your original idea was right.
Does probability of touch work the same for ES futures and SPX?
The math is the same, but the definition of touch differs. ES trades nearly 24 hours, so overnight prints count. SPX cash only prints during regular hours. Pick the series and session that match your orders.
Is a high probability of profit the same as a high win rate?
No. Probability of profit is a model estimate at entry. Your realized win rate also depends on how you manage the trade, because early exits and stops convert touches into outcomes the expiration math never counted.
Sources and further reading
- Probability and expectancy, options.wiki
- Probability of Touch glossary entry, SPXXL
- Probability of touching, Quantitative Finance Stack Exchange
- Two Ways to Analyze Expected Move, tastylive Market Measures
- 0DTE Oracle documentation, Option Alpha
This article is educational and describes general market concepts. It is not investment advice or a recommendation to buy or sell any security or contract. Options and futures trading, including 0DTE trading, carries a substantial risk of loss. Examples use fictional levels.
More from the blog
Gamma Exposure and 0DTE Pins: Why One Number Misreads the Day
One gamma exposure number hides the day. The gamma flip, positive vs negative gamma, pins vs max pain, and why 0DTE gamma sharpens into the close.
Dor S.Oct 26, 2026 · 12 min readAnchored VWAP and Liquidity: How to Read Where SPX and ES Agreed on Value
Learn how anchored VWAP (AVWAP) maps liquidity on ES and SPX: formulas, anchor selection, volatility, four readings and what options add.
Dor S.Oct 20, 2026 · 13 min read
