The flow that arrives when nothing happens
Open a chart on a quiet session — no data, no headline, the index unchanged at lunch — and size still trades. Part of that is the mechanical residue of two Greeks most options education never reaches.
Delta is not a fixed number. It drifts as the clock runs and as implied volatility moves, and desks that run delta-neutral books have to trade the underlying to keep pace with that drift whether or not they hold a view. Charm measures how delta changes as time passes. Vanna measures how delta changes as implied volatility moves.
Both sit one level above the core Greeks. They are cross-derivatives — the sensitivity of one Greek to a change in a different variable — and they matter for exactly the reason gamma matters. Positioning plus a second-order Greek equals a flow that has to happen.
What almost every explainer skips is that the direction of that flow is not a property of vanna or charm at all. It is a property of who holds what. That assumption is the thing to check before any of the rest is worth acting on.
Charm: delta on a clock
Charm is the rate of change of delta with respect to time — the partial derivative ∂Δ/∂t, sometimes called delta decay or delta bleed. Theta bleeds an option's value each day; charm bleeds its delta.
The behaviour splits by moneyness. An out-of-the-money option's delta decays toward zero as expiration nears, because the odds of finishing in the money shrink with every hour that fails to deliver a move. An in-the-money option's delta migrates the other way, toward 1.0 for a call or −1.0 for a put, as exercise becomes close to certain. Charm describes both migrations, and its magnitude scales inversely with time to expiration — a rounding error with sixty days left, the dominant force in the last two.
What makes charm distinctive is that it is scheduled. It needs no catalyst. Every day that passes moves deltas by an amount you could have computed a week earlier, and the hedges attached to them move with it.
This is the origin of the "charm bid." Institutions are structurally long index puts as portfolio insurance, which leaves dealers on the other side structurally short them. Short puts carry positive delta, so the desk sells the index to neutralise. As expiration approaches and those out-of-the-money put deltas decay toward zero, the short hedge is no longer needed and gets bought back. Steady, mechanical, opinion-free repurchasing into expiration.
Vanna: delta on a volatility dial
Vanna is the rate of change of delta with respect to implied volatility (∂Δ/∂σ). It is a cross-partial of the option's value with respect to spot and volatility, which means it reads two ways: vanna is equally the rate of change of vega with respect to the underlying price. Directional exposure moving with volatility and volatility exposure moving with price are the same number seen from two sides.
Where charm is the clock, vanna is the dial. When implied volatility rises, the deltas of out-of-the-money options increase in magnitude — a distant strike suddenly looks reachable, so it starts carrying directional weight. When implied volatility falls, those deltas shrink back toward zero. Cboe's VIX index is the standard reading of that variable for the S&P 500, described by the exchange as a leading measure of market expectations of near-term volatility conveyed by SPX option prices. Vanna is the channel through which a move in that measure becomes buying or selling in the index itself.
Take the same book of dealer-held short puts. If implied volatility falls on a calm, grinding session, those put deltas shrink, the short hedge is oversized, and the desk buys to cover. Falling volatility mechanically produces buying. If volatility spikes, put deltas grow, more of the underlying gets sold to re-hedge, and that selling compounds the decline. Part of the familiar negative correlation between price and volatility is not sentiment at all. It is hedge maintenance.
The sign lives in the positioning, not the Greek
Neither Greek generates a flow on its own. A Greek is a derivative; a flow needs an inventory to apply it to. This is where most vanna-and-charm commentary quietly breaks.
Everything above rests on one assumption — that dealers are net short index puts and net long gamma. That is the usual shape, and it follows from why customers buy options in the first place, as covered in the dealer positioning guide. But it is an inference, not an observation. Nobody publishes dealer inventory. It is reconstructed from open interest plus assumptions about which side initiated each trade, and it can be wrong, or simply stale by the time you read it.
Flip the assumption and every sign above flips with it. A book that is net long puts sells into charm decay rather than buying. The mechanism is unchanged; the output reverses. It is also why delta hedging is a continuous activity rather than a single trade: a desk that is flat at 9:30 a.m. is no longer flat by 3:30 p.m., because charm and vanna moved the target while it stood still.
The academic evidence that hedging flows move prices is strongest exactly where those flows are largest. Ni, Pearson and Poteshman documented in the Journal of Financial Economics that optionable stock prices cluster at strikes on expiration dates, finding that hedge rebalancing by option market makers — alongside trading by firm proprietary desks — contributed to the clustering. Expiration is precisely when charm is at maximum and the hedging requirement collapses fastest.
The OpEx cycle, and what 0DTE changed
The best-known application is calendar-based. Into a large monthly expiration, the charm bid and a typical drift lower in implied volatility stack on top of each other, and the resulting dealer buying is frequently associated with a slow grind higher. Flow desks including SpotGamma have documented this vanna and charm rally dynamic. Afterwards the expiring contracts and the hedges attached to them are simply gone, and the days that follow get called a "window of weakness" — not because selling appears, but because the mechanical buying that had been quietly supporting the tape stops.
Two mechanics are worth knowing before trading that calendar.
First, the monthly SPX contract does not stop trading when most people assume. Cboe's SPX specifications state that trading in the AM-settled contract "will ordinarily cease on the business day (usually a Thursday) preceding the day on which the exercise-settlement value (i.e., the expiration date) is calculated," while PM-settled SPXW options trade until 4:00 p.m. ET on the expiration day itself. The charm attached to the monthly tranche has already run out by Friday's open.
Second, the two-week framing comes from a market that no longer exists in the same proportions. Cboe reported that through the second quarter of 2026, 0DTE volume was up 46.2% year to date to more than 20 million contracts a day, with SPX 0DTE volume nearly tripling since the start of 2024. When a large share of contracts is created and extinguished inside a single session, charm runs a full cycle every day rather than once a month. The monthly effect still exists. It now competes with a daily one, and any build-into-OpEx narrative that ignores that is describing a smaller share of the flow than it used to.
When the gamma regime flips the script
There is one more conditional, and it is the one that turns a useful backdrop into a dangerous one.
Below the GEX flip level — where aggregate gamma exposure crosses into negative territory — dealers hedge with the move instead of against it, selling into weakness and buying into strength. In that regime a volatility spike does not merely produce vanna selling; it produces vanna selling that pushes price lower, which lifts implied volatility again. The stabilising loop becomes an amplifying one, and the charm bid weakens or reverses alongside it.
The practical consequence is that one headline carries two meanings. "Implied volatility is falling" is mechanical fuel in a positive-gamma regime and close to meaningless in a negative-gamma one. Locate price relative to the flip before drawing any conclusion from either Greek.
Reading them without a dealer desk
You will not compute these Greeks by hand, and you do not need to. What you need is the conditional structure, in order.
- Establish the gamma regime first. Positive gamma means stabilising flows and a credible charm bid; negative gamma means flows can amplify. SquawkFlow's free SPX GEX chart shows the flip level and the walls around it.
- Check the calendar, including which tranche is expiring and when it actually stops trading.
- Note the direction of implied volatility. Falling IV in positive gamma supports price through vanna; rising IV does the opposite.
- Watch the clock inside the day. Vanna repricing clusters in the first hour as overnight IV resets; charm drift concentrates into the close, especially before weekends and expirations.
- Treat all of it as backdrop. It describes the environment a move happens in, not whether a move is coming.
What vanna and charm do not tell you
They do not tell you direction. They describe whether a move, once started, gets damped or amplified — a conditional statement that needs a catalyst to condition on. A charm bid supports a tape right up until a real seller decides it does not care that dealers are buying.
They are also estimates built on estimates. The Greeks are exact given a model and a position; the position is inferred, the model is a simplification, and any aggregate published anywhere is somebody's reconstruction of a book nobody discloses. Positioning shifts intraday as new contracts trade, and the reconstruction lags it.
Used properly, they answer one narrow but genuinely useful question: is the tape currently being pushed by hedging, and is that push about to stop? That is worth knowing. It is not a thesis.
Educational content, not financial advice. See our risk disclosure.