Why ITM Spreads Trade Below Max Value

An ITM vertical spread trades below its max value on every day before expiration. A $5 wide spread sitting $20 in the money is a near certain $5 at expiry, and nobody will pay the full $5 for it today. The gap splits into four priced components, and each one is measurable, which turns hold or close into arithmetic instead of instinct.

What is the max value of a vertical spread?

A vertical spread is two options of the same type on the same underlying, same expiration, different strikes: one bought, one sold. The distance between the strikes is the width. Buy the 600 call, sell the 605 call, and the spread is $5 wide.

At expiration the arithmetic collapses. Finish above both strikes and the long leg exceeds the short leg by exactly the width: $5 a share, $500 a contract. Finish below both and the legs expire worthless together. The width is the ceiling, and that ceiling is what traders mean by max value. It is an expiration number, not a today number. Debit spreads and credit spreads share the same ceiling and differ in whether cash moved in or out when the position went on.

Why ITM spreads trade below max value

Before expiration you do not hold $5. You hold a claim on $5 with conditions attached. Four things live in the gap between what that claim trades at and the width:

• the time value of money, since $5 in 45 days is worth less than $5 in the account now
• the residual chance the underlying travels back through the strikes before expiry
• early assignment on the short leg, plus the cost of carrying the stock it delivers
• the bid to ask width you cross on the way out

The first three set fair value. The fourth sets what you can transact at.

A dollar at expiration is not a dollar today

A near certain $5 at a known future date is a zero coupon loan in an options costume: cash now, a fixed amount later. Price it the way a lender would. At a 5% annual rate with 45 days left, $5 divided by (1 + 0.05 × 45 / 365) is $4.9694, or $4.97 at the penny the market quotes in. Three cents of the gap is interest, before anything else enters. That is the same arithmetic behind a box spread's implied loan rate, where a package of options with a fixed payoff trades as a lending instrument. The greek that measures the sensitivity has a name: rho, the change in an option's price for a one point move in rates. The interest piece scales with the short rate, which has moved a long way in three years.

SELECT
    toString(toStartOfMonth(date))                     AS month,
    formatDateTime(toStartOfMonth(min(date)), '%b %Y') AS month_label,
    round(avg(toFloat64(yield_3_month)), 2)            AS yield_3_month_pct,
    round(avg(toFloat64(yield_1_year)), 2)             AS yield_1_year_pct
FROM global_markets.treasury_yields
WHERE date >= '2023-07-01'
  AND date <  '2026-07-01'
  AND yield_3_month > 0
GROUP BY month
ORDER BY month

The three month bill averaged 5.49% in Jul 2023 and 3.81% in Jun 2026. At those levels the interest inside a 45 day $5 spread is worth a few cents a contract. Near zero it would round away entirely, which is why this component is easy to ignore in a low rate era and impossible to ignore in this one. The one year yield in the same panel is the better reference for a spread expiring further out.

Does the gap close as expiration approaches?

It does, on both counts: fewer days left to discount, and less room for the underlying to travel back through the strikes.

Run the same hypothetical $5 payoff at a 5% annual rate across three horizons and the interest piece alone moves from about 6 cents at 90 days, to 3 cents at 45 days, to under a penny inside the final two weeks. That path is mechanical: shorten the day count and the discount shrinks with it.

The probability piece takes a steeper path. A deep in the money spread with three months to run has time for a move that puts one or both strikes back in play. The same spread with four sessions left needs that move to arrive almost immediately, and the price stops paying for a scenario the calendar has nearly closed off.

Convergence is uneven across contracts, which shows up in how the two pieces are distributed. A spread whose short strike sits a few dollars below the underlying keeps a live probability component into the final session, while one sitting far below converges toward the width from the interest side alone. Average the gap across a basket of contracts and it usually lands above the median gap, since a minority of pairs are still far from settled and they pull the mean.

How deep is deep enough?

Depth does most of the remaining work, and it works on one component only. Push the short strike far enough below the underlying and the residual chance of trading back through the strikes shrinks toward nothing. A $5 wide spread whose short strike sits $40 under the close needs a move most sessions do not deliver, and the price reflects that arithmetic in a smaller probability charge.

What depth cannot touch is the discount rate. The payoff is still a fixed amount at a fixed future date, and a fixed amount in the future is worth less than the same amount now at any positive rate. The gap flattens rather than falling to zero. The curve has a floor, and the floor is the interest piece from the section above, plus whatever the market prices for the short leg being taken away early.

Depth retires the probability component. It cannot retire the discount.

What early assignment on the short leg costs

Your short call is somebody else's long call, and an American style contract can be exercised on any business day. The common case is a dividend. A holder of a deep in the money call with almost no time value left can exercise the day before the ex dividend date and own the shares in time to be paid. Early assignment clusters around those dates, and the cash at stake is published a quarter ahead.

SELECT
    toString(ex_dividend_date)                    AS ex_date,
    formatDateTime(ex_dividend_date, '%b %e, %Y') AS ex_date_label,
    round(toFloat64(any(cash_amount)), 4)         AS dividend_usd
FROM global_markets.stocks_dividends
WHERE ticker = 'AAPL'
  AND ex_dividend_date >= '2022-01-01'
  AND ex_dividend_date <  '2026-07-01'
  AND cash_amount > 0
GROUP BY ex_dividend_date
ORDER BY ex_dividend_date

AAPL has gone ex dividend 18 times over the window above, most recently on May 11, 2026 at $0.27 a share. Assignment does not end the position. It converts the short call into short stock, held against a long call that is still alive, and that stock position carries financing costs plus the dividend owed across the ex date. The spread's mark prices that possibility in advance.

The closing mark is not a tradable price

Everything above is fair value. Exiting is a separate transaction, and the number your platform shows at the end of the day is not the number you trade at. Closing marks come from a procedure, commonly the midpoint of the quoted market at a stated time, with a fallback when the quote is stale; how option closing marks are set walks through the mechanics. A vertical has two legs, so the width you cross is two quoted markets rather than one. A penny of slippage per leg on a $5 wide spread is 2 cents off the exit, the same order of magnitude as the entire interest component at 45 days.

Hold or close, as arithmetic

Holding an in the money spread to expiration is a short dated lending decision. Capital stays tied up until expiry, and the return is the distance between the price you could exit at now and the width you collect later. Annualize that distance and it becomes comparable to cash.

Take a $5 wide spread quoted at $4.95 with 45 days to run. The remaining gain is 5 cents on $4.95 of committed capital, 1.01% over 45 days, or (5 / 4.95 - 1) × 365 / 45 = 8.19% a year. Set that beside the three month bill at 3.81% in Jun 2026: the hold is being paid a premium over cash for the residual travel risk and the assignment exposure. Run it again at a $4.99 quote and the annualized figure lands under 2%, beneath the bill. The arithmetic decides nothing. It states the price of waiting in units you can compare with something else.

FAQ

Does an ITM vertical spread ever reach max value before expiration?

Not in a tradable sense. The mark converges toward the width as the days run out and the underlying moves further through both strikes, and it can round to the width on a screen, but a bid at the full width would hand the buyer a fixed payoff with no compensation for the wait.

Why is my deep in the money call spread marked below the width?

Four pieces sit in the gap: the payoff is in the future and gets discounted at a short term rate, the underlying keeps some chance of trading back through the strikes, the short leg can be assigned early, and the bid to ask you cross shows up when you exit.

Can I lose money on an in the money spread I hold to expiration?

Yes. In the money describes today, not expiration day. If the underlying travels back below both strikes before expiry the spread expires worthless, and one that finishes between the strikes settles somewhere between zero and the width.

How do I annualize the yield left in an ITM spread?

Divide the width by the price you could exit at, subtract one, then scale by 365 divided by the days remaining. A $5 wide spread quoted at $4.95 with 45 days left works out to about 8.2% a year, before commissions and before any assignment on the short leg.

Every panel here ships with the exact SQL beneath it. To rebuild any of them on your own strikes, dates, or underlying, ask the question in plain English on the Strasmore terminal.