The Greeks: Delta, Gamma, Theta, Vega

Δ: Sensitivity to the Underlying

Δ is the partial derivative of the option price with respect to S, the underlying price. In plain reading: if S moves by $1, the option price moves by approximately Δ dollars, conditional on no other input changing. For a call, Δ is bounded in [0, +1]. For a put, in [−1, 0]. Those bounds are unconditional — they fall out of the no-arbitrage structure, not from any model.

Magnitude by Moneyness

Worked example. A call on AAPL with Δ = 0.60 trades at $3.00. AAPL rises $2. The first-order P&L estimate is $2 × 0.60 = $1.20, taking the option to roughly $4.20. If AAPL falls $2, the symmetric estimate is −$1.20. The word "approximately" matters: this is a first-order linearisation. The second-order correction is Γ, addressed below.

Δ as Risk-Neutral Probability

Under the Black-Scholes framework, Δ for a call equals N(d₁), which is closely related to but not identical to the risk-neutral probability of finishing in-the-money. The shorthand — "Δ = 0.30 means roughly a 30% chance of expiring ITM" — is useful as a back-of-envelope reading. Assuming log-normal returns and ignoring volatility skew, the approximation is reasonable. For real equity options, where skew is non-trivial and the distribution of returns is fatter-tailed than log-normal, the gap between Δ and the actual ITM probability is wider than the textbook suggests. Treat it as a rough indicator, not a calibrated probability.

Δ for Position Sizing

The most useful operational role of Δ is to convert option positions into share-equivalent exposure. Ten contracts × 100 multiplier × Δ 0.50 = 500 share-equivalents. A trader who has sized a $1,000 risk for a 100-share stock position with a $10 stop, and then buys 10 contracts of the same notional intent at Δ 0.50, has just put on 5x the intended exposure. The arithmetic is simple; the mistake is common. Divide intended share count by Δ to get contract count for matched directional exposure. See [[Position Sizing and Risk Management|position sizing]] for the broader framework.

Γ: The Convexity Term

Γ is the second partial derivative of the option price with respect to S, equivalently the first derivative of Δ with respect to S. It measures how fast Δ changes as the underlying moves. If Δ is the velocity of the option's response to S, Γ is the acceleration. Long-option positions are long Γ; short-option positions are short Γ. The asymmetry is the source of most of the interesting structure in options trading.

The Long-Γ Asymmetry

A call with Δ = 0.50 and Γ = 0.08 will, after a $1 rally in S, have a new Δ of approximately 0.58. The next $1 of upside therefore generates a larger dollar gain than the first. The position accelerates into your favour. After a $1 decline, Δ has fallen to roughly 0.42, so subsequent declines bleed less than the initial one. Gains accelerate; losses decelerate. That asymmetry is what you are paying for when you pay the option premium — and the daily cost of paying for it is Θ.

Where Γ Concentrates

Γ is highest for at-the-money options near expiration. The mathematical reason is that at the money, the probability is most sensitive to the next price tick — a small move shifts the option from "likely to expire worthless" to "likely to expire in-the-money," and Δ recalibrates sharply. Far from the money, the probability is already pinned near 0 or 1, and small moves do not change it much.

For sellers of options, the final week of expiration is the danger zone: short-Γ exposure goes vertical, which means small moves in S translate into large changes in mark-to-market P&L. Most experienced sellers of [[Covered Calls: Income From Your Portfolio|covered calls]] and other premium-collection structures close or roll positions before that final week, unless they specifically want the convexity exposure for a defined reason.

Θ: The Time-Decay Term

Θ is the partial derivative of the option price with respect to time T, expressed conventionally as a negative number for long options — meaning the option loses approximately |Θ| per calendar day, conditional on S and σ held constant. Time is a one-way variable. T only decreases. The premium associated with optionality bleeds out of the price as T approaches zero.

Daily Mechanics

Worked example. A long call worth $4.50 with Θ = −0.08. Tomorrow, holding everything else fixed, the model price is $4.42. The day after, $4.34. Over 30 calendar days, the cumulative theta bill is roughly $2.40 — over half the option's current value, paid in daily instalments to hold the position.

The Decay Curve Is Not Linear

Θ itself is not constant; it accelerates as T → 0. The functional form, for an at-the-money option in the Black-Scholes framework, is approximately proportional to 1/√T, which is gentle when T is large and steep when T is small. Concretely:

• 60 DTE: Θ modest. ATM options lose roughly $0.03–0.05 per day.
• 30 DTE: Daily decay roughly doubles relative to 60 DTE.
• 14 DTE: Decay is 3–4× the 60-DTE rate.
• Final week: ATM options can lose 5–10% of remaining value per calendar day.

The 1/√T shape is why the operational rule of thumb — buy at 30–60 DTE, sell at 30 DTE or less — tends to hold up. Buyers want time for the directional thesis to play out without paying the steep portion of the curve. Sellers want to harvest the steep portion. The covered-call structure is, at its core, a Θ trade.

Weekend Theta

Pricing models use calendar days, not trading days. The Friday close already reflects the model's expected weekend decay. The empirical observation that options "barely move" over weekends is the consequence: Thursday and Friday afternoon prints have absorbed most of it. Some traders sell options on Friday in an attempt to harvest a calendar-mechanics edge, but market makers know the model as well, and the marks are calibrated. Modulo wide bid-ask spreads on illiquid names, the edge has been priced in.

ν (Vega): Sensitivity to Implied Volatility

ν is the partial derivative of the option price with respect to σ, the implied volatility input. Note that ν (vega) is not, strictly, a Greek letter — it is the conventional symbol assigned to fill the slot. The convention is universal in trading practice. By convention, ν is quoted as the dollar change in option price per 1-percentage-point change in σ.

What Implied Volatility Is

Implied volatility is the σ that, when plugged into the Black-Scholes formula along with the four observable inputs (S, K, T, r), reproduces the observed market price of the option. It is therefore a rearrangement of the equation, not a measurement of the underlying's actual volatility. The market's IV is a forecast embedded in option prices; whether the underlying realises that σ over the life of the contract is a separate empirical question. Usually it does not match exactly; the gap between implied σ and realised σ is most of the tradeable structure in options markets (Sharpe-adjusted, the spread between IV and subsequent realised vol tightens substantially — the math is in any standard options textbook).

Worked Example

Long call worth $5.00, ν = 0.15, current IV = 25%. If IV rises to 27%, the first-order estimate of the new option price is $5.00 + 0.15 × 2 = $5.30. If IV falls to 22%, $5.00 − 0.15 × 3 = $4.55. S has not moved in either case; the entire P&L is the volatility input shifting.

This is why options can lose money on directional trades that "worked." Buying AAPL calls into earnings, S rallies $4 the next morning, and IV collapses from 60% to 30% as the event uncertainty resolves. The Δ gain on the rally and the ν loss on the IV collapse are competing forces, and on event days the ν term often dominates. Traders call this IV crush. It is the canonical way that beginners lose money on earnings plays, even when they are right on direction. See [[Earnings Season: How to Trade Around Reports|earnings season]] for the broader trading-around-events framing.

ν Scales with Time to Expiry

Longer-dated options have higher ν than shorter-dated ones — roughly proportional to √T in the Black-Scholes framework. A 12-month LEAPS might have ν of 0.40; a weekly might have ν of 0.03. If the trade is "I think IV is too low and will rise," structure with longer-dated options to maximise ν exposure. If the trade is "I think IV is too high and will fall," sell longer-dated options.

The Greeks Acting Jointly

None of the Greeks acts in isolation. The first-order Taylor expansion of the option price change over a single day is, approximately:

ΔV ≈ Δ · ΔS + ½ Γ · (ΔS)² + Θ · Δt + ν · Δσ

Scenario. Long 10 AAPL calls, 30 DTE, ATM. Per-contract values: Δ = 0.55, Γ = 0.04, Θ = −0.06, ν = 0.12. Multiplier 100. Position aggregates: 5.50 Δ (550 share-equivalents long), 0.40 Γ, −$60/day Θ, 1,200 ν-cents per IV point.

Day 1: S rises $1.50, IV falls 1 point.

• Δ term: 550 × $1.50 = +$825
• Γ term: ½ × 0.04 × 100 × 10 × 1.50² ≈ +$45
• Θ term: −$60
• ν term: 1,200 × (−$0.01) = −$12
• Net ≈ +$798

Direction dominated. Θ and ν ran against the position and were absorbed by the move. This is the long-option ideal.

Day 2: S flat, IV falls another 2 points.

• Δ term: $0
• Γ term: 0 (no ΔS)
• Θ term: −$60
• ν term: 1,200 × (−$0.02) = −$24
• Net ≈ −$84

No directional move, no Γ harvest. Θ and ν are the only active terms, both negative. This is the long-option failure mode: a flat tape with declining IV bleeds the position daily, and the bleed is paid on the schedule of the calendar regardless of intraday optimism.

Strategy by Greek Profile

Directional Conviction

If the thesis is directional, the cleanest expression is to maximise Δ and minimise ν and Θ sensitivity. Buy ITM options at Δ 0.70–0.80. They behave nearly like the underlying with limited downside, low Γ, low ν, and modest Θ per dollar of premium. Treat them as a stock substitute with a built-in stop. See [[Risk/Reward Ratio: The Math Behind Every Trade|risk/reward]] for sizing.

Volatility Conviction

If the thesis is "a large move is coming, direction unspecified," the position to construct is long Γ and long ν — an ATM straddle or strangle. The cost is Θ: the position pays a daily fee to hold the gamma-and-vega exposure. Pre-earnings is the canonical setup, with the standard caveat that ν risk runs in two directions on event days. IV crush is the typical failure mode.

Premium Collection

If the thesis is "a quiet or slowly drifting tape," the position to construct is short Θ-positive structures — out-of-the-money [[Covered Calls: Income From Your Portfolio|covered calls]], cash-secured puts, or credit spreads — with 30 DTE or less. The position collects time decay on a daily schedule. The risk is Γ: a sharp adverse move in S can produce a loss exceeding many days of accumulated Θ. Strict position sizing is non-negotiable for this style.

Common Errors

• Buying cheap OTM options without computing Θ as a percentage of premium. A $0.50 option with Θ = −0.03 is bleeding 6% per day. The required move-and-pace in the underlying to break even is steeper than it looks at the screen.
• Buying into earnings without modelling IV crush. The forward IV (post-event) is typically much lower than the IV paid going in. The break-even on S has to clear both the strike distance and the ν loss from the IV collapse. Many event trades that "should have worked" did not, for this reason.
• Holding short positions through the final week of expiration. Γ concentrates there. The mark-to-market sensitivity is non-linear in S, and sizing that was comfortable at 30 DTE is uncomfortable at 3 DTE.
• Treating Δ as fixed. A trade plan that assumes Δ stays at 0.50 across a $5 move in S is mis-specified by the Γ term. The realised P&L will diverge from the linear projection in proportion to ½Γ(ΔS)².

The Greeks turn options from opaque instruments into measurable exposures. For any position under consideration, four numbers should be on the page before the order is sent: Δ (directional exposure), Γ (convexity), Θ (daily carrying cost), ν (volatility exposure). If any of those four cannot be quantified for the position, the position is not fully understood.

Assumptions, to be explicit for this article: Black-Scholes framework, log-normal returns, continuous hedging, no transaction costs, constant r, constant σ within each first-order step. Real markets violate every one of those, in varying degrees. Fat tails make Γ risk asymmetric in practice. Discrete rebalancing introduces hedge-error variance. Bid-ask spreads make every Greek-implied P&L a best-case mark. Expect 1-in-30 days to break the linear projection.

See also: [[Options Trading: Calls, Puts, and the Basics|Options Basics]] · [[Covered Calls: Income From Your Portfolio|Covered Calls]] · [[Position Sizing and Risk Management]] · [[Risk/Reward Ratio: The Math Behind Every Trade|Risk/Reward]] · [[Earnings Season: How to Trade Around Reports|Earnings Season]]