The short answer
Theta and charm both answer "what does one more day do to this option." They answer it about different things.
- Theta is the change in the option's value per day. Units: currency per contract per day. It shows up in a profit and loss statement.
- Charm is the change in the option's delta per day. Units: delta per day, and once aggregated across a chain, dollars of index delta per day. It shows up in somebody's order flow.
A position can be bleeding theta while its charm is near zero, and it can be carrying large charm on a day where theta is a rounding error. They are separate derivatives of the same price with respect to the same variable, taken one level apart.
The four core Greeks covers theta itself, and vanna and charm explained covers charm as a concept and why the flow it produces depends on positioning. This page is only about the comparison.
Where each one is largest
This is the difference that catches people, because the two peak in different places on the chain.
Theta is largest at the money. An at-the-money option holds the most time value, so it has the most to lose per day. Move away from the money in either direction and the absolute theta falls, because there is less optionality left to decay.
Charm is near zero at the money. A forward at-the-money contract has a delta near 0.5 in magnitude, positive for a call and negative for a put, and keeps it as the clock runs, because the odds of finishing above or below stay balanced. There is almost nothing for time to move. Charm is largest some distance from the money, where delta is actively migrating: toward zero for a contract the clock is pushing out of the money, and toward its in-the-money limit for one it is locking in, which is plus one for a call and minus one for a put.
The distance is not vague. At the zero-rate convention this site computes on, charm is the standard normal density at d1 multiplied by d2 and divided by twice the time remaining, which is the same expression for a call and for a put at that strike. The size of that product peaks roughly one standard deviation of remaining move either side of the forward, and it falls away in both directions from there. What the two sides do not share is the sign, which follows the sign of d2: strikes above the forward carry negative charm and strikes below it carry positive charm, for calls and puts alike. Above the forward a call's delta drains toward zero and a put's toward minus one, so both fall; below it a call's delta climbs toward one and a put's toward zero, so both rise. The exact crossing, where d2 is zero, sits a hair below the forward.
So a chain on expiration morning can show its biggest theta numbers and its biggest charm numbers at different strikes, and both readings are correct.
Where they behave alike
Both accelerate as expiration approaches, and both stop.
Theta for an at-the-money option, where theta is largest, scales roughly as one over the square root of time remaining. The largest charm on the chain, which sits about one standard deviation of remaining move from the money, scales roughly as one over the time remaining, so it steepens faster. With spot at 7,780 and 11 percent vol, Black-Scholes at-the-money theta is about 15.5 times larger at three hours to expiry than at thirty days, the square root of the 240 to 1 ratio between the two clocks. Over the same window the largest charm on the ladder, run through the function this site computes with, grows about 233 times, close to the full 240. Among strikes near spot, the one struck exactly at the money grows slowest, about as slowly as theta, and it is also where charm is close to zero; strikes 5 to 20 points away grow more than a thousand times over the same window. Strikes more than about four three-hour standard deviations out, some 65 points here, see their charm shrink rather than grow, because by then their delta has almost nothing left to resolve. Both Greeks are rounding errors with sixty days left and dominant in the final session.
Both also run out. A contract whose delta has finished resolving has no more delta for time to move, and a contract with no time value left has none to decay. Neither Greek rises indefinitely into the bell. On the SPX chain captured at 1:33pm Eastern on 2026-09-22, net charm across the whole book ran at about minus 43.3 billion dollars of dealer delta per day, and by the 4:00pm mark on the same capture it ran at about minus 17.1 billion, with the same-day component at zero once those contracts had no time left.
Why the distinction matters outside a P&L
Theta is a cost carried by whoever is long the option and a credit to whoever is short it. It is settled between the two of them and produces no trade in the underlying.
Charm does produce a trade in the underlying. A desk running a delta-neutral book has a delta target of zero, and if the deltas of the options it holds drift overnight and through the session, the hedge has to move to match. That is why charm appears in positioning data at all and theta does not: one is an accounting outcome, the other is an order.
That is also why the aggregate versions exist on different terms. Vanna exposure and charm exposure sums charm across the chain and quotes it in dollars of dealer index delta per day, and the live SPX figure sits in the second-order exposure section under the free gamma levels. There is no equivalent aggregate theta figure published here, because a summed theta would describe a transfer between counterparties rather than a flow into the index.
The three confusions worth naming
"Theta decay moved my delta." It did not. Theta moved the premium. The delta change over the same period is charm, and on a position some distance from the money the two can be large at once and unrelated in size.
"Charm is just theta for delta, so the signs match." They do not have to. Theta is negative for almost every long option. Charm is not: its sign flips with moneyness, through the sign of d2, and it is the same sign for a call and a put at the same strike. Put-call parity is the quickest way to see that. Delta on a call minus delta on a put equals exp(-qT), a quantity the clock moves only at the rate of the dividend yield, so the two shed delta at essentially the same rate, and at the zero-yield convention the figures on this site use, at exactly the same rate. What makes an aggregate charm figure come out positive or negative is not the right but the assumed dealer side: long the calls and short the puts, as set out below, so two contracts with identical charm enter the sum with opposite signs, and the total is that netting across strikes above and below spot.
"Both are biggest on expiration day, so the last hour is the peak for each." Both are biggest on expiration day. Neither peaks at the bell. Theta on a worthless out-of-the-money contract is finished before the close, and so is the charm on a delta that has already gone to zero. Why the market moves in the last hour works through what that does to the shape of the afternoon.
What neither one tells you
Neither Greek says what price will do. A Greek is a sensitivity, and turning it into a flow requires an inventory to apply it to. The sign on every aggregate charm figure published anywhere, including here, rests on the convention that dealers are long calls and short puts against customer flow. Open interest shows that a contract exists. It never shows which side a dealer holds.
Educational content, not financial advice. See our risk disclosure.